1

Governors Island in New York Bay - Formation of the Island and of Some

Historical Archaeological Contexts

Donald M. Thieme, Department of Physics, Astronomy, and Geosciences, Valdosta State University,

1500 North Patterson Street, Valdosta, Georgia, 31698

Manuscript accepted for publication in the University of Florida Press volume, Geoarchaeology in Historical and Urban Contexts, edited by A. Gilbert and J. Schuldenrein

May 19, 2008

Governors Island is a special place within the New York City metropolis, with a strategic

location in New York Bay that has led to military presence on the island throughout much of

American history. The island's standing architecture will attract visitors and new residents for

many years to come, while beneath their feet remain many historic artifacts and other residues of

past human activity. Recent archaeological survey and testing provided an opportunity to employ

current earth science methods as well as to compare the new findings with what was already

known about the most recent geological period, the Quaternary, in southern New York.

In addition to confirming several previous hypotheses regarding the formation of

Governors Island, the present study identifies sediment packages transported onto the island’s

surface by specific geological processes. There is some bedrock, particularly beneath the

southern end of the island. Above the bedrock is glacial till, a poorly sorted mixture of sand,

clay, gravel, and boulders that was deposited in close proximity to glaciers toward the end of the

Pleistocene epoch of the Quaternary period. The till rarely extends to the surface, however, and

2

the intact sediment packages that are immediately below the island's surface are all relatively

coarse sand consisting almost entirely of the mineral quartz. These "cover sand" packages

contain some significant archaeological contexts and show considerable variation in texture and

chemistry depending upon how they were transported to the island, how long they sat beneath

forest vegetation on the island, and how much they were disturbed by human activity since the

first Americans of European descent arrived at the island.

At the southeast and northwest edges of the original island, particularly well-sorted sand

deposits are found in the vicinity of old ("paleo") shorelines. Remnants of these paleoshorelines

are also shown on historic maps, some 150 meters inland of the current position of the active

shore (Garman and Herbster, 1996). These paleoshorelines promise to yield data that can be

compared with the rising sea levels throughout the eastern United States during the past 10,000

years, the Holocene epoch of the geological time scale (Harland et al., 1990).

Toward the top of the original island and adjacent to the large star-shaped fort, Fort Jay,

there are "midden" deposits rich in food refuse as well as 19th century historic artifacts. Whereas

the midden deposits are concentrated in areas that extend for less than ten meters in any

direction, a more extensive package of relatively fine-textured and organic-rich sediment extends

across most of the original island and beneath all of the historical archaeological contexts. This is

described here as a "paleosol" or soil that was formed on a landscape of the past (Birkeland,

1999, p. 339-346). Because of the extensive landscaping associated with Fort Jay and other

buildings on the island, the paleosol is found at varying depths below the present land surface

and has also been removed or disturbed in places by those landscaping activities. Radiometric

dates and other laboratory results obtained in the present study support the hypothesis that the

3

paleosol formed at the top of a land surface that preceded the first arrival to the island of

Americans of European descent.

Methodology - Geology of Historical Archaeological Contexts

Contributions of earth science figure prominently in the archaeology of Early Man sites

(e.g. Farrand, 1975, 2001; Hay, 1976, 1990) and of deeply stratified prehistoric sites (Donahue

and Adovasio, 1990; Farrand, 1993; Haynes and Agogino, 1966; Larsen and Schuldenrein, 1990;

Waters, 1986). Less attention is typically paid to geological processes by archaeologists who

excavate sites that contain buildings which are shown on historic maps or mentioned in written

documents. An emphasis on deliberate human design is perhaps to be expected given the

significant contributions by architectural and archival research to historical archaeology. From a

methodological perspective, however, the excavation and analysis of any archaeological context

should be as complete as possible, both in obtaining data from the finds and in explaining those

data on the basis of scientific theories. While it may be true that geological processes have less

effect on the provenance of artifacts and the composition of cultural sediments from the most

recent portion of the archaeological record, this must nonetheless be determined from detailed

analysis of archaeological contexts. Analysis of the contexts encountered during archaeological

survey and test excavations on Governors Island in New York Bay provides data that further this

research objective as well as substantiating several previous hypotheses regarding the formation

of the island itself.

In studying the formation of the island and its archaeological contexts, geological and

archaeological research problems were inseparably intertwined. Scientific methods from both

geology and archaeology were therefore applied in a synthetic manner in order to understand

4

deposits produced by processes that fall on a continuum between "natural" and "cultural" end-

members (Butzer, 1982; Schiffer, 1987; Stein, 2001). The field investigations themselves

involved a collaboration where archaeologists troweled floors, sieved sediments, and identified

cultural features within machine-excavated geological trenches (Herbster et al., 1997; Garman et

al., 1998). Archaeological test units excavated entirely by hand were also described and sampled

by the project geologist. Archaeological analysis of artifacts included as particles in the island's

deposits helped to constrain the age of those deposits beyond what could have been achieved

through stratigraphy and radiometric dating alone.

Most of the data for the present study come from sixteen small (1 x 5 m) trenches which

were excavated mechanically in order to describe and sample the stratigraphy of the original

island. Three of these trenches identified paleoshorelines at the southeast and northwest edges of

the island. Due to the addition of landfill and other engineering modifications, the

paleoshorelines occur up to 150 meters inland of the active shore. Although they are not within

the Governors Island National Monument as presently designated, these locations have yielded

some data that are relevant to reconstructing local and regional environmental change (e.g.

Fairbridge and Newman, 1968; Salwen, 1962; Weiss, 1974).

Historical archaeological contexts are particularly rich repositories of stratigraphic detail

from the past few centuries of the Quaternary period. However, the questions asked by earth

scientists are not typically at the right "scale" for the large volume of deposits associated with

relatively brief episodes of human activity on archaeological sites (Stein, 1991). In our studies on

Governors Island, geological questions about the formation of the island were at one end of a

scale continuum, with the opposite end represented by archaeological questions about the

formation and "stratigraphic integrity" of the island's archaeological contexts.

5

Figure 1: Location of Governors Island in New York Bay

6

These archaeological research questions were addressed during both the initial "Phase IB"

survey (Herbster et al., 1997) and a subsequent "Phase II" testing phase (Garman et al., 1998) by

describing and sampling the profiles of excavation units which exposed midden and other

contexts from which historic artifacts were recovered. Both the contribution of residues from

various human activities and the "integrity" of the layers in which those residues are found were

examined using a nested series of laboratory analyses.

Based upon both the stratigraphy and the laboratory results, the island's "cultural"

sediments can be distinguished from the relatively coarse sands dominated by quartz which were

deposited when the island initially formed at the end of the last ice age. Sediments that are

younger than these late Pleistocene sands but older than the first arrival on the island of

Americans of European descent were also encountered in a few locations. In particular, an

organic-rich deposit identified as a buried paleosol was found to contain sparse lithic debris. This

late prehistoric paleosol toward the top of the original island has been intruded and disturbed by

historic activity and now extends beneath Fort Jay within the Governors Island National

Monument. Preservation and development plans by both the National Park Service (2007) and

the Governors Island Preservation and Education Corporation (2007) promise to protect the

island's cultural resources and continue to fund scientific research that involves both geologists

and archaeologists.

Study Area and Regional Stratigraphic Framework

Governors Island is less than a kilometer south of the southern tip of Manhattan in New

York Bay (see Figure 1). The island presently totals nearly 70 hectares (ha), less than 40 ha of

which represent the "original" island that existed prior to the beginning of the 20th century

7

Figure 2: Cross-section of New York Bay prepared from borings for the Brooklyn-Battery Tunnel (Sandborn, 1950)

8

(Garman and Herbster, 1996; Langan, 2005). Between 1902 and 1911, the southwest end was

extended through the addition of material that was removed from Manhattan during the

construction of the Lexington Avenue subway. The Governors Island National Monument, the

focus of the present study, consists of a mere 8.9 ha surrounding a complex of historic buildings

built on the original, geological island (Figure 2). The entire island, both original and artificial, is

currently administered by the Governors Island Preservation and Education Corporation (2007).

The most detailed previous description of the bedrock and glacial deposits in this portion

of New York Bay comes from studies performed for the Brooklyn-Battery Tunnel (Sanborn,

1950; Schuberth, 1968). The tunnel runs along the northeast side of the island, and a large

ventilator shaft for the tunnel can be easily seen from the island shore. On the cross-sections

prepared from tunnel borings (Sanborn, 1950), bedrock of "Manhattan schist" is shown beneath

less than 15 m of unconsolidated sediment (Figure 2). Although Governors Island cannot be

called a bedrock island, schist bedrock does outcrop very near the surface. Schist rock rich in

muscovite, a clear or yellowish-brown mineral of the mica group (Dietrich and Skinner, 1979, p.

47-49), was encountered at a depth of two meters below the land surface (2 m bls) in two of the

trenches excavated for the present study.

The deposits at the present land surface on Governors Island were mapped by Fullerton et

al. (1992) with a unit that includes "lake, ice contact, and outwash deposits," while Cadwell

(1989) mapped them as "till." Several layers of till, a poorly sorted mixture of sand, clay, gravel,

and boulders deposited immediately in front of a glacial ice sheet, are shown in the Brooklyn-

Battery Tunnel cross-sections, primarily in areas north of the ventilator shaft. The till layers are

from three to five meters thick, and at least one till was identified in two of the trenches

excavated on the island itself in the present study. As described in these trenches (MT-9 and MT-

9

13), the till contains clasts of granite, diabase, gneiss, and other erratics in a matrix of yellowish

brown sand and clay.

In both of the tunnel cross-sections, tills alternate with "drift" (stratified sand and gravel).

Stratified sand, silt and clay, probably deposited on the bottom and shores of a glacial lake, are

shown primarily on the south side of the island, extending beneath the Buttermilk Channel. Two

borings performed for the U. S. Army Corps of Engineers prior to dredging of the Buttermilk

Channel (GRA, 2000) also identified stratified sand, silt, and clay. The stratification is in the

form of thin "varves" only a few centimeters thick which are inferred to represent annual layers

of deposition. Both above and below the varved lake sediment were layers of glacial till that

included organic mud. A bulk sample of the organic mud from 4.6 m below the channel floor in

one of the borings was submitted for radiocarbon dating to Beta Analytic of Coral Gables,

Florida, returning an age of 26,000±300 years before present ("BP"). The material dated was

overlain by coarse, poorly sorted sand interbedded with brown clayey silt. This is probably a

"flow till" (Hartshorn, 1958) of late Wisconsinan age. Varved silt and clay found at

approximately the same depth in the other boring suggests that the ice was melting directly into

the lake at this location. Gray-green, sandy "summer" varves alternate with reddish brown

"winter" muds in the lake sediment.

No varved sediments were identified in the trenches on the island itself. Sand interbedded

with the glacial till does have the same reddish brown color found in the glacial lake mud. The

color comes from the Newark group sandstones and mudstones of the present Hackensack and

Passaic drainages in northern New Jersey (Schuberth, 1968; Stanford and Harper, 1991). The

sediments between the top of the glacial till or stratified drift and the present land surface on

Governors Island are mostly sand, coarser in texture than the estuarine mud at the top of borings

10

in New York Bay (GRA, 2000; Sanborn, 1950; Weiss, 1974). As noted above, textural and

geochemical variation in these "cover sands" on the island were found to be related to how they

were transported to the island, how long they sat beneath forest vegetation on the island, and how

much they were disturbed by human activity. All of the archaeological contexts on Governors

Island are associated with the cover sands or with the paleosol formed in these sands.

Stratigraphy and Archaeology of the Original Island

Prior to the beginning of the 20th century, Governors Island was much smaller,

amounting to only approximately 40 hectares of its present total of 70 ha (Garman and Herbster,

1996; Langan, 2005). Five areas shown to have been part of this original island on historic maps

and photographs were investigated with stratigraphic trenches during archaeological survey

("Phase IB") and testing ("Phase II") by the Public Archaeology Laboratory (Herbster et al.,

1997; Garman et al., 1998). The distribution of trenches within the five areas is summarized in

Table 1, and the trench locations are plotted in Figure 3. In addition to the stratigraphic trenches

excavated using heavy equipment, several smaller units which the archaeologists excavated with

hand tools during the Phase II study were described and sampled as part of the geoarchaeological

study.

Linking the profiles for trenches MT-1, MT-16, MT-12B, and MT-13(6), we have a

cross-section running from south to north across the original portion of Governors Island (Figure

4). A cross-section running from west to east has been prepared by linking the profiles for

trenches MT-7, MT-9, MT-14, MT-5, and one of the archaeological excavation units (EU-5) in

Nolan Park (Figure 5). Figure 4 shows that schist bedrock was reached in MT-1 by coring with a

hand auger through angular regolith. The same bedrock was also encountered below stratified

11

Table 1: Distribution of Stratigraphic Trenches within Survey Areas on Governors Island in New York Bay

Symbol Name Stratigraphic Trenches

(B) Fort Jay MT-4, MT-12A, MT-12B

(G) South Battery MT-1

(H) Nolan Park MT-5, MT-19

(I) Golf Course MT-1, MT-6, MT-13, MT-14, MT-16, MT-17

(J-1) Colonel's Row MT-3, MT-8, MT-9

(J-2) Hay Road MT-7

Figure 3: Map of the original portion of Governors Island showing the locations of stratigraphic trenches

12

Table 2: Radiocarbon Dates obtained for samples collected during the archaeological investigations on Governors Island

Survey Area Trench Horizon

14C yr BP Cal Intercept Lab No.

B Fort Jay MT-12A 2Ab 590±60 AD 1400 Beta-117732

H Nolan Park STP A 460±60 AD 1570-1630 Beta-117725

I Golf Course MT-16 II 660±70 AD 1295-1400 Beta-117733

I Golf Course MT-6 2Ab 1130±110 AD 785-1015 Beta-107658

J Colonel's Row MT-9 2BC 2610±50 BC 800 Beta-107659

Figure 4: Cross-section running north-south across the original portion of Governors Island

13

Figure 5: Cross-section running east-west across the original portion of Governors Island

glacial deposits approximately 130 m to the west at the base of trench MT-3. The occurrence of

Manhattan schist at relatively shallow depths (2-3 m bls) in the southern portion of the original

island is consistent with the cross-sections for the Brooklyn-Battery Tunnel (Figure 2). The

bedrock is folded up to the south in this part of New York Bay, and this resulted in the

emergence of relatively resistant rock as sea level fell during the last glaciation. The bedrock

core acted as a "snag" for glacial debris and other sandy detritus, building the island to the north

(up ice).

14

Glacial till on Governors Island could either be an isolated drumlin feature or part of a

more extensive recessional moraine that extends beneath New York Bay (Cadwell, 1989;

Fullerton et al., 1992; Sirkin, 1986; Stanford, 1997). The till is thickest just west of Fort Jay (see

Figure 5), where it was reached within the upper 1.5 m of excavation in trenches MT-14 and

MT-9. The till contains clasts of granite, diabase, gneiss, and other erratics in a matrix of

yellowish brown sand and clay. Several sets were found interbedded with reddish brown

(5YR5/4) sand in an excavation that reached a total depth of over 5 m bls at trench MT-14. The

elevation at the top of the till slopes steeply westward between the MT-14 and MT-9 locations at

approximately the same grade as the land surface, and this natural relief was evidently exploited

for the construction of the western side of Fort Jay. In the MT-9 profile, a relatively well-

developed paleosol was found to have formed in sandy outwash that capped the till. A

radiocarbon date of 2610±50 BP was obtained on the paleosol sediment, collected at 80-90 cm

bls (~3 m MSL) in trench MT-9 (see Table 2).

West of the MT-9 location on the Figure 5 cross-section is the MT-7 location where the

surface organic mat (Ah horizon) rests immediately on top of sediments that were deposited

along a paleoshoreline. The sediments deposited in the "shoreface" depositional environment at

this location are well-sorted sand and consist primarily of the mineral quartz (quartzose)

interbedded with a "hash" of bivalve and gastropod shell fragments. Examined at a scale of a few

centimeters, the sand shows ripple bedding caused by waves. The shell hash is found directly

above the ripple beds, which have then been truncated by erosion. Sedimentologists refer to this

as a "reactivation surface" (Friedman and Sanders, 1978, p. 542-543; Prothero and Schwab,

2004, p. 49-50), found primarily in an intertidal setting where the wave energy is occurring

during only one tidal phase.

15

Beds deeper in the trench MT-7 profile (60-80 cm bls) contain whole shells of the eastern

oyster (Crassostrea virginica). These beds probably relate to a different tidal phase or to a

depositional environment dominated by tides rather than waves (Cooper, 2007, p. 271-274). At

100 cm bls in trench MT-7, beneath these laminar beds, was another reactivation surface marked

by a lag of fine shell hash. A hand auger boring in the trench floor showed that laminated sand

and sandy mud occur at least 220 cm bls (~ 1 m MSL). The sandy mud represents an estuary

environment relatively undisturbed by wave energy. A local relative sea level rise appears to

have occurred between the time that this depositional environment was in effect and the

deposition of the wave-rippled sands associated with the paleoshoreline.

The trenches in the berm around Fort Jay (Area B) and on the Golf Course (Area I)

typically did not reach intact natural sediments of the original island within the upper meter.

Human activities subsequent to the purchase of the island by the Dutch in 1637 have here formed

a complex sequence which represents the cultural end-member of depositional environments.

Construction fill, demolition rubble, and refuse middens are among the deposits present.

Radiocarbon dates can only identify these as "modern," so young that their radiocarbon

abundance is within the range of the peaks produced by fossil fuel combustion (Aitken, 1990, p.

66-72; Suess, 1955). Ceramics, glass, and other artifacts recovered by the archaeologists provide

more definitive age constraints, and time-sensitive artifacts are particularly abundant in the

deposits beneath the Fort Jay berm (MT-4, MT-12A, and MT-12B) and in Nolan Park (EU-5,

EU-7). In addition to the archaeological data, maps and other documents record the construction

of Fort Jay (1806-1809), Castle Williams (1807-1811), and other historic structures on the island

(NPS, 2003).

16

Traces of Native American cultural activity on Governors Island are more sparse, and the

prehistoric ceramic and lithic artifacts recovered by the Public Archaeology Laboratory were all

from contexts that lack stratigraphic integrity (Herbster et al., 1997). There is potential for better

preservation of prehistoric contexts in deposits that underlie the historic fill and refuse on both

the north and the south sides of Fort Jay. Radiocarbon dates of 1130±110 BP, 660±70 BP, and

590±60 BP were obtained on scattered wood charcoal from a paleosol found both north of the

fort at MT-13(6) and south of the fort at MT-4, MT-12A, MT-12B, and MT-16. Particle size and

geochemistry of the sediments from this paleosol are consistent with seasonal settlement by

Native Americans during the Middle Woodland and Late Woodland periods of regional

prehistory (Ritchie, 1969).

Particle Size Analysis and Sedimentology

Particle size analysis provides key data for reconstructing environments of deposition for

archaeological sediments as well as for water-borne or windborne particles (Leigh, 2001; Stein,

2001). In the present study, the percent sand, silt, and clay were determined by hydrometer

(Bouyoucos, 1962) for a total of 29 sediment samples from both archaeological contexts and the

intact deposits of the original portion of Governors Island. Sand-sized particles make up the bulk

of all of the samples studied, but the sediment from archaeological contexts is much finer

textured than that from the basal cover sand. The contrast can be clearly seen in a ternary plot

comparing samples from middens and pit features in trenches MT-12A, MT-12B, MT-13, MT-

16, and two of the archaeological excavation units (EU-5 and EU-7) with samples of basal sand

from trenches MT-2, MT-5, MT-7, MT-8, MT-12B, MT-13, and MT-16 (Figure 6). The basal

17

Figure 6: Ternary plot (lower half) comparing the particle size of samples from archaeological contexts with samples of basal cover sand, Governors Island

Figure 7: Sieve analysis results for 14 samples of basal cover sand from Governors Island

18

("C" horizon) sand samples are over 90% sand, whereas the archaeological samples all have

either sandy loam or loam texture (Birkeland, 1999; Soil Survey Staff, 1997, 2006).

The amount of silt and clay in the archaeological deposits is probably within the range

that would be found in a natural topsoil formed under deciduous forest on the island. Soil series

recently mapped on Staten Island, for example, have loam or sandy loam textures in the A and B

horizons (Hernandez and Galbraith, 1997). The abundance of artifacts and evidence for land

modification in all of the deposits investigated on the original island prevented a controlled

comparison with natural topsoil. In addition to the artifacts found and the stratigraphic

relationships that suggest land modification, however, some of the physical and chemical

attributes of the archaeological deposits are diagnostic for that type of deposit. Much of what

altered the parent material to make the sediments of these archaeological contexts appears to

have been the result of the changing cultural environment on the island.

Fourteen of the 29 samples in the particle size study were from the basal "cover sand"

deposits on Governors Island. These basal sands contain mica and other weatherable minerals, in

addition to quartz. Mica is particularly abundant in the glacial deposits and in deposits that

appear to have been subjected to river (fluvial) transport. River transport is indicated by the

results of a sieve analysis using a one "phi" interval (Anderson, 2007; Burt, 2004; Day, 1965;

Gee and Bauder, 1986). Five of the fourteen basal cover sand samples analyzed had a distinctive

"fine tail" of abundant silt and clay (Folk, 1966; Friedman, 1961, 1967). These are the five with

over 20% in phi sizes greater than four ("> 4") on Figure 7 and represent basal sediments from

trenches MT-3, MT-12A, MT-12B, and MT-13 as well as the 2BC horizon in MT-12A.

The distribution of particles by phi size for basal sands from Governors Island points to a

number of specific depositional environments, not all of which were immediately evident from

19

Figure 8: Plot of skewness versus sorting for 14 samples of basal cover sand from Governors Island

the stratigraphy. Figure 8 plots skewness versus sorting for the fourteen samples. The sorting

index used is a simplified form of the standard deviation (Friedman and Sanders, 1978, p. 75).

Better sorted samples have lower values and plot to the left. The samples with a fluvial "fine tail"

are very poorly sorted and positively skewed, plotting in the upper right and inferred to be from a

glaciofluvial environment.

20

All but one of the glaciofluvial samples are from the trenches north and south of Fort Jay,

flanking the large till deposit. They underlie contexts from which late Holocene radiocarbon

dates were obtained. The channel in this area was either contemporaneous with the till or slightly

younger. Another outwash deposit occurred near the base of trench MT-3, 200 m south of Fort

Jay in an area that was later submerged by late Holocene sea level rise. The MT-3 sample is

extremely poorly sorted, with five percent gravel as well as abundant silt and clay (25%).

The remaining nine samples are moderately sorted, and these are the samples with the

high peaks from 1-4 phi in Figure 7. Winnowing of fines by wave action could have occurred

either on a bay shore similar to the present (fluviomarine) or on the shore of a glacial lake during

the late Pleistocene (glaciolacustrine). The samples from MT-13 and MT-14 are deep in the

stratigraphy and probably late Pleistocene. The samples from MT-5, MT-7, and MT-8 are from

Holocene paleoshorelines. Two samples were actually sieved from MT-7, one of which is

negatively skewed as is typical of an ocean beach in a shoreface depositional environment

(Friedman, 1967). Glaciofluvial, fluviomarine, shoreface, and possibly glaciolacustrine

depositional energies are thus indicated by particle size distributions in samples of basal sand

from Governors Island. The samples from archaeological contexts have more silt and clay than

occur in the basal sands, and this is because they contain fines that were contributed by human

activity.

Geochemistry of the Archaeological Contexts

Chemical analysis of sediment samples has been used in previous studies to identify

archaeological sites, to evaluate their stratigraphic integrity, and even to distinguish residues

from particular types of human activity (Barba et al., 1996; Entwistle et al., 1998; Foss et al.,

21

1994a, 1994b; Kolb et al., 1990; Middleton and Price, 1996; Schuldenrein, 1995). A "nested"

approach was used in the present study. Preliminary identification of samples that contain fine-

textured anthropogenic residues was based on the particle size results and measurements of soil

pH and percent organic matter for 21 of the 29 samples in the study. The pH was measured on a

slurry prepared with a solution of 0.01 N CaCl2 (McClean, 1982; Peters, 2007). Organic matter

content was measured by loss-on-ignition (Dean, 1974; Stein, 1984).

Available phosphorus and exchangeable Ca, Mg, and K were measured for sixteen

samples from contexts whose formation could be reconstructed independently on the basis of

stratigraphy and included artifacts. The extraction was done with 1 M ammonium acetate

followed by atomic absorption spectrophotometry (Peters, 2007). Although the procedure is

designed to estimate the availability of nutrients for crop plants, it has also been used to identify

anthropogenic residues in previous studies (Kolb et al., 1990; Schuldenrein, 1995).

The abundance of Pb and Zn was measured for ten of the samples to evaluate the utility

of these two elements in distinguishing residues from specific activities. Atomic absorption

spectrophotometry was used once again, but the extraction was done with DTPA

(diethylenetriaminepentaacetic acid). Finally, total phosphate and the relative contribution by the

three phosphate fractions defined by Eidt (1977, 1985) were measured for five samples from

specific archaeological contexts around the perimeter of Fort Jay and in Nolan Park.

Organic matter enrichment associated with human activity is evident in a plot comparing

percent silt and clay with percent organic matter for the 21 samples analyzed (Figure 9).

Contexts containing fine-textured anthropogenic residues all plot in the upper right of the figure,

with at least 30% silt- and clay-sized particles and at least one percent organic matter. The

contexts are in three locations. There is a sample from the paleosol which underlies deep fill

22

Figure 9: Percent organic matter versus percent silt and clay for 21 samples from the Governors Island geoarchaeological study

deposits on the north side of Fort Jay at trench MT-13(6). Prehistoric lithics, marine shell, and

historic glass and ceramics were collected from the trench profile, although the paleosol has not

yet been sampled by an archaeological excavation. On the south side of Fort Jay (Area B), there

is a historic midden which is also deeply buried beneath the berm and in turn rests upon a

paleosol that represents the intact late prehistoric land surface. Prehistoric ceramics as well as

historic ceramics, glass, buttons, and even a gunflint were recovered from the archaeological test

excavations within the three trenches located here (MT-4, MT-12A, and MT-12B). The trench

23

MT-16 location is on the Golf Course (Area I), approximately ten meters south of the Fort Jay

berm trenches and beyond the limits of the midden. The samples from MT-16 can therefore serve

as controls against which to compare the samples containing anthropogenic residues.

The third location where fine-textured anthropogenic residues appear to occur in

association with both prehistoric and historic artifacts is Nolan Park (Area H). Prehistoric

ceramic sherds and chipping debris from stone tool manufacture were found at the location

sampled by archaeological excavation unit 5 (EU-5). Unfortunately, the excavation of EU-5 did

not yield cultural pit features or artifact concentrations that indicate a prehistoric living surface.

The geochemical results for the column of six sediment samples collected from the north wall of

EU-5 therefore provide key information about the contexts in which prehistoric artifacts occur in

Nolan Park. The sediment texture, percent organic matter, and phosphate fraction amounts were

also measured for a sample from a historic feature at the base of another archaeological

excavation unit in Nolan Park, EU-7. The results tend to confirm the contribution of

anthropogenic residues to this sample, as can be seen from Figure 9.

Very large amounts of available phosphorus (>200 mg/kg) in the upper three samples

from EU-5 suggest that phosphate was added in recent decades to fertilize the grass in Nolan

Park. The samples from the historic midden and the paleosol north and south of Fort Jay have

from 50 to 100 mg/kg available phosphorus (see Figure 10). These values fall within the range

for topsoil formed under forest vegetation in the eastern United States (Kleinman et al., 2002;

Scott et al., 2001). Residual phosphate is thus preserved within the fine fraction of the deeply

buried prehistoric and early historic sediments on Governors Island.

The total phosphate was between 250 and 500 mg/kg for the five phosphate fractionation

samples (Figure 11), a range that is characteristic of residues from dwelling, gardening,

24

Figure 10: Available phosphorus (mg/kg) versus percent silt and clay for 16 samples from the Governors Island geoarchaeological study

25

Figure 11: Ternary plot of phosphate fractionation results for five samples from archaeological contexts on Governors Island

26

Table 3: Chemistry of MT-12B Profile, Fort Jay Berm, Governors Island

Horizon Depth bls OM pH K Mg P Ca Pb Zn

Midden 150 cm 1.5 4.9 6 29 82.7 307.0 255.1 0.34

2Ab 200 cm 0.5 5.2 4 30 33.0 150.0 0.7 0.10

2C1 250 cm 0.4 5.3 5 28 46.5 168.0 0.1 0.12

2C2 300 cm 0.7 5.2 - - - - - - - - - - - -

Table 4: Chemistry of EU-5 Soil Profile in Nolan Park, Governors Island

Horizon Depth bls OM pH K Mg P Ca Pb Zn

A1 5 cm 6.3 4.9 48 89 426.5 700.0

A2 15 cm 4.7 4.3 44 38 424.2 233.0 354.6 0.12

AB1 25 cm 2.4 4.3 36 19 250.1 109.0

AB2 40 cm 2.4 4.2 32 12 107.1 87.0 68.6 0.10

BC1 55 cm 1.3 4.2 23 16 28.4 89.0

BC2 65 cm 1.1 4.1 33 19 12.6 95.0 14.3 0.06

27

manufacturing, and refuse disposal activities (Eidt, 1977; Schuldenrein, 1995). Fraction I,

consisting of loosely bound iron and aluminum phosphates, accounts for over half of the total in

the samples from trench MT-13, from the AB2 horizon in Nolan Park EU-5, and from the

historic post feature in Nolan Park EU-7. The samples from the historic midden and the paleosol

in trench MT-12B have much higher loadings on Fraction III (calcium-bound phosphates) but

lower loadings on Fraction II (occluded iron and aluminum phosphates). There are two likely

sources for calcium-bound phosphates in these contexts, lime mortar in the case of the historic

midden and bone or marine shell in the case of both the midden and the paleosol.

The chemistry of samples collected from beneath the Fort Jay berm at MT-12B (Table 3)

indicates discrete, relatively rapid episodes of deposition which produced contexts containing

sediments with very different physical and chemical properties. In addition to organic matter and

phosphate, calcium, lead, and zinc occur in the historic midden in greater amounts than in the

underlying paleosol (2Ab) or its parent material (2C1). Because the midden sediment is actually

slightly more acid, with a pH of 4.9 compared to 5.3 for the 2C1 horizon (Table 3), it is unlikely

that these elements were precipitated after deposition. Previous studies have found calcium and

zinc to be abundant in refuse that contains human excrement, fish bones, and shell (Kolb et al.,

1990; Middleton and Price, 1996; Schuldenrein, 1995; Scudder et al., 1995). Lead levels greater

than 50 mg/kg are common in residues from societies that practice metallurgy (Foss et al., 1994a,

1994b; Macklin et al., 1992). Lead therefore seems to be particularly suited to distinguishing

residues from historic as opposed to prehistoric cultural activity in North America. In addition,

lead may be particularly abundant in residues from military garrison areas where lead shot was

stored and weapons were cleaned.

28

Soil chemistry changes much more gradually between the samples collected from EU-5

in Nolan Park. This suggests that here deposition was so gradual that soil development kept up

the same pace, mixing the sediment to produce gradual transitions in physical and chemical

properties between horizons (Table 4). The Pb values fall off smoothly down profile in EU-5

from over 350 mg/kg in the A1 horizon. There is still 14 mg/kg in the BC2 horizon at 65 cm

below surface, an amount which is over twice that in any of the other samples from prehistoric

contexts or culturally sterile deposits on the island. The available P falls off smoothly like the Pb

while Ca, Mg, and K increase from the AB2 to the BC2 horizon. Lead has moved down profile

at least 50 cm since it was first deposited at the land surface. Smooth transitions down profile in

all of the elements that were measured indicate that the Nolan Park deposits have been

repeatedly mixed by biological organisms and other disturbance as they built up to the level of

the present land surface.

Summary of Governors Island Geoarchaeology

Results of earth science investigations have contributed significantly to our knowledge of

Governors Island and its archaeological contexts. Due to the artificial extension of the island to

the southwest between 1902 and 1911 (Garman and Herbster, 1996; Langan, 2005), contexts

over 100 years old are found only on the original, northerly portion of the island. Artifacts and

other cultural residues occur there in sediments which overlie and intrude late Quaternary "cover

sand" deposits. By sieving fourteen samples of these sands at one phi (Anderson, 2007; Burt,

2004; Day, 1965; Gee and Bauder, 1986), it has been possible to sketch out the complex history

of glacial, fluvial, lacustrine, and marine depositional energies which built the island. Additional

analyses of particle size for samples from the archaeological contexts show a characteristic

29

abundance of fine material, and chemical analyses further identify some specific human

activities that may be contributing that material.

Results of the particle size analysis indicate that there was lateral variation in the initial

texture and chemistry of sediments on the island. Glacial till and drift or outwash deposition

appear to be responsible for the tail of fine material found in samples from trenches MT-12A,

MT-12B, and MT-13, which plot in the upper right corner of Figure 8. A particularly high and

thick deposit of till and stratified drift toward the north end of Governors Island appears to have

been chosen as the location of Fort Jay. This location would have already offered natural

protection as a fighting position and required minimal earth moving during initial construction of

the fort between 1806 and 1809. Relatively coarse sands were later used to construct the berm

and other earthen architectural features.

Even in locations found to be devoid of both historic and prehistoric artifacts, the cover

sand deposits on Governors Island preserve much unique information about the changing natural

environment of New York Bay. Deposits associated with paleoshorelines located at the southeast

and northwest edges of the original island, in particular, appear to record a local sea level

highstand previously identified in New York Bay by Fairbridge (1960a, 1960b, 1961) and others

(e.g. Salwen, 1962). On the northwest edge of the island, the maximum age of the paleoshoreline

has been constrained by a radiocarbon date of 2610±50 B.P. for paleosol sediment. In terms of

particle size, samples from trenches MT-5, MT-7, and MT-8 typify shoreface deposition with a

fine sand to medium sand texture and low index values for sorting (moderate or moderately well

sorted). One of the MT-7 samples was also negatively skewed, toward the coarse end of its size

distribution, which is very typical of an ocean beach in a shoreface depositional environment

(Friedman, 1967).

30

Both Schiffer (1987) and Stein (2001) have emphasized that natural depositional and

postdepositional processes always govern the formation of archaeological contexts, even when

people are immediately responsible and even when they document what they have done as is true

for many historical sites. For heuristic purposes, it is useful to think of some deposits as more

"cultural" and others as more "natural" along a continuum with two extreme end members

(Butzer, 1982). The "cultural" sediments that contain artifacts usually do differ in terms of both

particle size and geochemistry, as was found in the Governors Island contexts described above.

As has been argued by Stein (1985 p. 7; 2001, p. 21), anthropogenic or "cultural"

sediments should always be studied through controlled comparison with sediments from purely

natural, "geogenic" deposits. On Governors Island, however, there really are no "offsite"

depositional contexts due to extensive disturbance for more than 300 years. The approach taken

here is actually one suggested by Stein (1985, p. 7; 2001, p. 21) in which control samples are

obtained from strata within an archaeological excavation that are devoid of cultural material.

Samples from midden, a possible post feature, and topsoil that contains historic and prehistoric

artifacts were studied through comparison to "cover sand" C horizon samples, many of which

came from directly beneath archaeological contexts.

Another approach for controlled comparison is to make use of the regional

sedimentological and soil geochemical databases which have now been developed for much of

the United States. Chemical signatures identified in the present study fall above or in the upper

range of reported values for soil series mapped on Staten Island (Hernandez and Galbraith,

1997). In the case of phosphorus, there are also systematic studies of soils formed under forest

vegetation in the eastern United States (Kleinman et al., 2002; Scott et al., 2001). The topsoil

31

samples from Nolan Park do appear to be aberrant, possibly due to recent addition of artificial

fertilizer.

In the contexts where fine-textured residues were introduced by human activities, the

sediment characteristics result from a combination of geological and cultural processes.

Chemical results indicate that human excreta have incremented the percent organic matter, the

available phosphorous, the loosely bound iron and aluminum phosphates, and possibly the trace

amounts of zinc in sediments from archaeological contexts. Higher zinc values also appear to

occur in sediments to which marine shell, animal bone, and other kitchen refuse were added, and

these sediments also have abundant exchangeable calcium and calcium bound phosphates.

Storing and firing lead shot as well as maintaining military weapons have increased the amount

of lead in 19th and 20th century deposits in the original portion of Governors Island. Because lead

is only abundant in residues from societies that practice metallurgy (Foss et al., 1994a, 1994b;

Macklin et al., 1992), it seems to be particularly suited to distinguishing residues from historic as

opposed to prehistoric cultural activity in North America. Values greater than 250 mg/kg, as

found in two of the sediment samples in the present study. probably occur only in soils from

specific cultural environments, such as the military garrison areas on Governors Island.

The identification of chemical residues associated with specific human activities is useful

to archaeologists interpreting a context that contains prehistoric or historic artifacts. At least as

important from a management or planning perspective, however, is the use of earth science

methods including chemical analysis to assess the stratigraphic integrity of archaeological

contexts. On Governors Island, contexts associated with Fort Jay and other features of 19th and

20th century construction and landscaping are easily identified from abrupt changes in the color

and texture of associated sediments. In other cases, however, artifact-rich middens and dwelling

32

areas have been mixed with topsoil material and buried up to four meters below the present land

surface. These more deeply buried contexts are challenging to identify and interpret, because

they were formed by both natural and cultural processes (Schiffer, 1987; Stein, 2001)

On Governors Island as in other areas mantled by late Quaternary cover sands in the

Atlantic Coastal Plain (Leigh, 2001; Michie, 1990; Scudder et al., 1996), physical mixing by

small organisms ("pedoturbation") and chemical weathering are important natural processes

affecting archaeological contexts and their surrounding matrix. Beyond simply identifying the

presence or absence of anthropogenic residues, sedimentological and geochemical results can be

used to assess the stratigraphic integrity of archaeological contexts in such deposits. Profiles

containing sediments with very different physical and chemical properties, such as that beneath

the Fort Jay berm at trench MT-12B, indicate discrete, relatively rapid episodes of deposition.

Profiles where deposition was so gradual that soil development kept up the same pace, such as

that in Nolan Park EU-5, are considered to be "disturbed" and to lack stratigraphic integrity for

archaeological purposes. Prehistoric and historic artifacts and residues need not be entirely

mixed within such a profile, however. The Nolan Park contexts do preserve some traces of

Middle Woodland and Late Woodland occupations in this portion of the original island.

The Governors Island study illustrates many of the challenges of doing geoarchaeology

on historical archaeological sites in that the problems investigated by earth scientists studying the

Quaternary period in southern New York are not an immediate concern of the archaeologists

who work on historical sites. Geological problems about processes responsible for sediment

deposition on the island do become archaeologically relevant, however, when framed as

questions about the formation and stratigraphic integrity of the island's archaeological contexts.

As noted by Stein (1991), geoarchaeology is very often a matter of bringing earth science

33

descriptions and interpretations to the right "scale" for the large volume of deposits associated

with relatively brief episodes of human activity on archaeological sites. In a similar discussion

referring specifically to chronology, Stewart (1999) has recently commended the archaeological

focus on a "human" scale of centuries up to millennia, bridging the gap between the documentary

archives of historians and the geological archives of Quaternary stratigraphers.

Due in part to the difficulties of scale mentioned above and in part to the management of

the island's resources by several government agencies, a number of questions and concerns

remain to be addressed by further geoarchaeological investigations on Governors Island. In order

to enrich and extend the robust results obtained through chemical analysis of the sediments from

archaeological contexts, for example, it would be helpful to take some samples of 20th century

contexts to analyze in comparison to the values of lead, phosphorus, and other elements in the

18th and 19th century archaeological contexts. In several areas, particularly on the Golf Course

and in Nolan Park, geophysical methods promise to provide information that compliments and

enriches the interpretations based upon stratigraphic trenching. In the paleoshoreline locations at

the southeast and northwest edge of the original island, it should be possible to more precisely

constrain the age of regional sea level highstands by dating shell in the deposits with radiocarbon

(Little, 1995) and by using luminescence methods to date the sand grains themselves (Aitken,

1998; Forman et al., 1992; Wallinga et al., 2001). Overall, the results from "Phase I" and "Phase

II" summarized above represent a successful synthesis of scientific methods from both geology

and archaeology, leading to a new understanding of how Governors Island formed and how

people contributed to its present form and the composition of its uppermost sediments.

34

REFERENCES CITED Aitken, Martin J., 1990, Science-based Dating in Archaeology. London: Longmans. Aitken, Martin J., 1998, An Introduction to Optical Dating. Oxford: Oxford University Press. Anderson, John R., 2007, Laboratory 7, Sand Sieve Analysis. In Historical Geology Laboratory Manual, edited by Pamela Gore. Atlanta, Ga.: Georgia Perimeter College. <http://www.gpc.edu/~pgore/geology/historical_lab/contents.php>, accessed June 1, 2007. Barba, Luis A., Ortiz, Agustin, Link, Karl F., Luján, Leonardo Lopez, and Lazos, Luz, 1996, Chemical Analysis of Residues in Floors and the Reconstruction of Ritual Activities at the Templo Mayor, Mexico. In Archaeological Chemistry - Organic, Inorganic, and Biochemical Analysis, edited by N. V. Orna, pp. 139-156. Washington, D. C.: American Chemical Society . Birkeland, Peter W.,1999, Soils and Geomorphology. Oxford: Oxford University Press. Bouyoucos, George J., 1962, Hydrometer method improved for making particle size analyses of soils. Agronomy Journal 54: 464-465. Burt, Rebecca, editor, 2004, Soil Survey Laboratory Methods Manual. Washington, D.C.: Natural Resources Conservation Service, U. S. Department of Agriculture. <http://soils.usda.gov/technical/lmm/>, accessed June 1, 2007. Butzer, Karl W., 1982, Archaeology as human ecology. Cambridge: Cambridge University Press. Cadwell, Donald H., 1989, Surficial Geologic Map of New York, Lower Hudson Sheet. Albany: New York State Museum. Cooper, Andrew, 2007, Temperate coastal environments. In Environmental Sedimentology, edited by C. Perry and K. Taylor, pp. 263-301. Oxford: Blackwell. Day, Paul R., 1965, Particle Fractionation and Particle-Size Analysis. In Methods of Soil Analysis, edited by Charles A. Black, pp. 545-567. Madison: Soil Science Society of America. Dean, Walter E. Jr., 1984, Determination of carbonate and organic matter in in calcareous sediment and sedimentary rocks by loss on ignition - Comparison with other methods. Journal of Sedimentary Petrology 44: 242-248. Dietrich, Richard V., and Skinner, Brian J., 1979, Rocks and Rock Minerals. New York: John Wiley and Sons. Donahue, Jack, and Adovasio, James M., 1990, Evolution of sandstone rockshelters in eastern North America: A geoarchaeological perspective. In Archaeological Geology of North America, edited by N. P. Lasca and J. Donahue, pp. 231-251. Boulder: Geological Society of America. Eidt, Robert C., 1977, Detection and Examination of Anthrosols by Phosphate Analysis. Science 197: 1327-1333. Eidt, Robert C., 1985, Theoretical and practical considerations in the analysis of anthrosols. In Archaeological Geology, edited by G. Rapp, Jr., pp. 215-229. New Haven: Yale University Press. Entwistle, J. A., Abrahams, P. W., and Dadgshon, R. A., Multi-Elemental Analysis of Soils from Scottish Historical Sites. Interpreting Land-use History Through the Physical and Geochemical Analysis of Soil. Journal of Archaeological Science 25: 53-68. Fairbridge, Rhodes W., 1960a, The Changing Level of the Sea. Scientific American, 202(5): 70-79.

35

Fairbridge, Rhodes W., 1960b, Recent World-Wide Sea Level Changes. Bulletin of the Massachusetts Archaeological Society 21, nos. 3-4: 49-51. Fairbridge, Rhodes W., 1961, Eustatic Changes in Sea Level. In Physics and Chemistry of the Earth, vol. 4, edited by L. H. Ahrens, F. Press, K. Rankama, and S. K. Runcorn, pp. 99-185. New York: Pergamon Press. Fairbridge, Rhodes W., and Newman, Walter S., 1968, Postglacial subsidence of the New York area. Zeitschrift fur Geomorphologie 3: 296-317. Farrand, William R., 1975, Sediment Analysis of a Prehistoric Rockshelter: The Abri Pataud. Quaternary Research 5: 1026. Farrand, William R., 1993, Discontinuity in the Stratigraphic Record. In Formation Processes in Archaeological Context, edited by Paul Goldberg, David T. Nash, and Michael D. Petraglia, pp. 85-96. Madison: Prehistory Press. Farrand, William R., 2001, Archaeological Sediments in Rockshelters and Caves. In Sediments in Archaeological Context, edited by J. K. Stein and W. R. Farrand, pp. 29-66. Salt Lake City: University of Utah Press. Folk, Robert L., 1966, A review of grain-size parameters. Sedimentology 6: 73-93.

Forman, Steven L., Bettis, E. Arthur III, Kemmis, Timothy J. and Miller, Barry B., 1992, Chronologic evidence for multiple periods of loess deposition during the late Pleistocene in the Missouri and Mississippi Valley, United States: Implications for the activity of the Laurentide ice sheet. Paleogeography, Paleoclimatology, Paleoecology 93: 71-83.

Foss, John E., Lewis, R. J., Stiles, Cynthia A., Lee, S. Y., and Timpson, Michael E., 1994a, Evidence of environmental contamination at several Roman archaeological sites in Italy and Tunisia. In Environmental Contamination, 6th International Conference, p. 189-191. Edinburgh: CEP Consultants. Foss, John E., Timpson, Michael E. Timpson, Lewis, R. J., and Stiles, Cynthia A., 1994b, Soil interpretation of some Roman archaeological sites in Italy and Tunisia. Proceedings of the 15th International Congress of Soil Scientists 66: 172-173. Friedman, Gerald M., 1961, Distinction between dune, beach, and river sands from their textural characteristics. Journal of Sedimentary Petrology 31, no. 4: 514-529. Friedman, Gerald M., 1967, Dynamic processes and statistical parameters compared for size frequency distribution of beach and river sands. Journal of Sedimentary Petrology 37, no. 2: 327-354. Friedman, G. M., and Sanders, J. E., 1978, Principles of Sedimentology. New York: John Wiley and Sons. Fullerton, Donald S., Sevon, W. D., Muller, E. H., Judson, S., Black, R. F., Wagner, P. W., Hartshorn, J. H., Chapman, W. F., and Cowan, W. D., 1992, Quaternary Map of the Hudson River 4o x 6o Quadrangle, United States and Canada. Washington, D. C.: U. S. Geological Survey, Miscellaneous Investigations Series. Garman, James C., and Herbster, Holly, 1996, Phase IA Archaeological Assessment of the Governors Island National Historic Landmark District, Governors Island, New York. The Public Archaeology Laboratory, Inc. Report no. 751, submitted to ABB Environmental Services, Inc. and the United States Coast Guard/Civil Engineering Unit, Warwick, Rhode Island Garman, James C., Russo, Paul A., Schuldenrein, Joseph, and Thieme, Donald M., 1998, Technical Report, Phase II Evaluation of the Archaeological Sites in the Governors Island National Historic Landmark District, Governors Island, New York. Public Archaeology Laboratory, Inc., Warwick, Rhode Island. Gee, Glendon W., and Bauder, James W., 1986, Particle-size analysis. In Methods of Soil Analysis, Part 1. Physical and mineralogical methods, pp. 383-411. Madison: Soil Science Society of America.

36

Geoarcheology Research Associates (GRA), 2000, Geomorphological and Archeological Study of New York and New Jersey Harbor Navigation Channels. Report prepared for the U. S. Army Corps of Engineers, New York District. Governors Island Preservation and Education Corporation (GIPEC), 2007, Governors Island maps and planning documents <http://www.govisland.com>, accessed June 1, 2007. Harland, Walter B., Armstrong, Richard L., Cox, Allen V., Craig, Lorraine E., Smith, Alan G., and Smith, David G., 1990, A geologic time scale. Cambridge: Cambridge University Press. Hartshorn, Joseph H., 1958, Flowtill in Southeastern Massachusetts. Geological Society of America Bulletin 69: 477-482. Hay, Richard L., 1976, Geology of Olduvai Gorge. Berkeley: University of California Press. Hay, Richard L., 1990, Olduvai Gorge: Interpretation of hominid paleoenvironments. In Establishment of a Geologic Framework for Paleoanthropology, edited by L. F. Laporte, pp. 23-37. Boulder, Co.: Geological Society of America Special Paper 242. Haynes, C. Vance, and Agogino, George A., 1966, Prehistoric springs and geochronology of the Clovis site, New Mexico. American Antiquity 31: 812-821. Herbster, Holly, Garman, James C., Schuldenrein, Joseph, and Thieme, Donald, 1997, Phase IB Archaeological Survey of the Governors Island National Historic Landmark District, Governor Island, New York. The Public Archaeology Laboratory, Inc. Report no. 851, submitted to HRP Associates, Inc. and the United States Coast Guard/Civil Engineering Unit, Warwick, Rhode Island.

Hernandez, L. A., and Galbraith, J. M, 1997, Soil Survey of South Latourette Park, Staten Island, New York City, NY. Washington, D.C.: U. S. Department of Agriculture, Natural Resources Conservation Service. Kleinman, Peter J., Sharpley, Andrew N., Moyer, Barton G., and Elwinger, Gerald F., Effect of Mineral and Manure Phosphorus Sources on Runoff Phosphorus. Journal of Environmental Quality 3: 2026-2033.

Kolb, Michael F., Lasca, Norman P., and Goldstein, Lynne G., 1990m A soil-geomorphic analysis of the midden deposits at the Aztalan site, Wisconsin. In Archaeological Geology of North America, edited by N. P. Lasca and J. Donahue, pp. 199-218. Boulder: Geological Society of America.

Langan Engineering and Environmental Services (Langan), 2005, Geotechnical Historic Data Report, Governors Island Development, New York, New York. Available upon request from the Governors Island Preservation and Education Corporation <http://www.govisland.com>, accessed June 1, 2007. Larsen, Curtis E., and Schuldenrein, Joseph, 1990, Depositional history of an archaeologically dated flood plain, Haw River, North Carolina. In Archaeological Geology of North America, edited by N. P. Lasca and J. Donahue, pp. 161-181. Boulder: Geological Society of America. Leigh, David S., 2001, Buried Artifacts in Sandy Soils: Techniques for Evaluating Pedoturbation versus Sedimentation. In Earth Sciences and Archaeology, edited by P. Goldberg, V. T. Holliday, and C. Reid Ferring, pp. 269-293. New York: Kluwer Academic/Plenum Publishers. Little, Elizabeth, 1995, Apples and Oranges - Radiocarbon Dates on Shell and Charcoal at Dogan Point. In Dogan Point - A Shell Matrix Site in the Lower Hudson Valley, edited by Cheryl Claasen, pp.121-128. Bethlehem: Archaeological Services.

37

Macklin, Mark G., Passmore, D. G., and Rumbsby, B. T., 1992, Climatic and cultural signals in Holocene alluvial sequences: the Tyne basin, northern England. In Alluvial Archaeology in Britain, edited by Stuart Needham and Mark G. Macklin, pp. 123-139. Oxford: Oxbow Books. McLean, E. O., 1982, Soil pH and lime requirement. In Methods of Soil Analysis, Part 2. Chemical and microbiological methods, pp. 119-223. Madison: Soil Science Society of America. Michie, James, 1990, Bioturbation and Gravity as a Potential Site Formation Process: The Open Area Site, 38GE261, Georgetown County, South Carolina. South Carolina Antiquities 22: 27-46. Middleton, William D. and Price, T. Douglas, 1996, Identification of Activity Areas by Multi-element Characterization of Sediments from Modern and Archaeological House Floors using Inductively Coupled Plasma-Atomic Emission Spectroscopy. Journal of Archaeological Science 23: 673-687. National Park Service, 2003, Proclamation 7647 - Establishment of the Governors Island National Monument. Federal Register 68, no. 28: 7053-7055. National Park Service, 2007, Personal communication regarding preservation planning and salvage excavations on Governors Island in New York Bay <http://www.nps.gov/gois/>, accessed June 1, 2007. Peters, John, 2007, Wisconsin Procedures for Soil Testing, Plant Analysis, and Feed and Forage Analysis. Madison: Soil and Plant Analysis Lab, University of Wisconsin. <http://uwlab.soils.wisc.edu/madison/>, accessed June 1, 2007. Prothero, Donald P., and Schwab, Fred, 2004, Sedimentary Geology - An Introduction to Sedimentary Rocks and Stratigraphy. New York: W. H. Freeman and Company. Ritchie, William A., 1969, The Archaeology of New York State. New York: Natural History Press. Salwen, Bert, 1962, Sea Levels and Archaeology in the Long Island Sound Area, American Antiquity, 28, no. 1: 46-55. Sanborn, James F., 1950, Engineering Geology in the design and construction of tunnels. In Application of Geology to Engineering Practice, edited by S. Paige, pp. 45-82. Boulder: Geological Society of America. Schiffer, Michael B., 1987, Formation Processes of the Archaeological Record. Albuquerque: University of New Mexico Press. Schuberth, Christopher J., 1968, The Geology of New York City and Environs. Garden City: Natural History Press. Schuldenrein, Joseph, 1995, Geochemistry, Phosphate Fractionation, and the Detection of Activity Areas at Prehistoric North American Sites. In Pedological Perspectives in Archaeological Research, edited by M. E. Collins, B. J. Carter, B. G. Gladfelter, and R. J. Southard, pp. 107-132. Madison: Soil Science Society of America. Scott, Christopher A., Walter, Michael F., Nagle, Gregory N., Walter, M. Todd, Sierra, Natalie V., and Brooks, Erin S., 2001, Residual phosphorus in runoff from successional forest on abandoned agricultural land: 1. Biogeochemical and hydrological processes. Biogeochemistry 55: 293-309. Scudder, Sylvia J., Foss, John E., and Collins, Mary E., 1996, Soil science in archaeology, Advances in Agronomy 57: pp. 1–76. Sirkin, Les, 1986, Pleistocene Stratigraphy of Long Island, New York. In The Wisconsinan Stage of the First Geological District, Eastern New York, edited by D. H. Cadwell, pp. 6-21. Albany: New York State Museum Bulletin Number 455. Soil Survey Staff, 1997, Keys to Soil Taxonomy. Blacksburg: Pocahontas Press.

38

Soil Survey Staff, 2006, Keys to Soil Taxonomy. Washington, D.C.: National Resources Conservation Service, U. S. Department of Agriculture. <http://soils.usda.gov/technical/ classification/tax_keys/>, accessed June 1, 2007. Stanford, Scott D., 1997, Pliocene-Quaternary geology of northern New Jersey - An overview. In Pliocene-Quaternary Geology of Northern New Jersey, Guideblook for the 60th Annual Reunion of the Northeastern Friends of the Pleistocene, pp. 1-1 to 1-26. Trenton: New Jersey Geological Survey. Stanford, Scott D., and David P. Harper, 1991, Glacial lakes of the lower Passaic, Hackensack, and lower Hudson valley, New Jersey and New York. Northeastern Geology 13: 277-286. Stein, Julie K., 1984, Organic Matter and Carbonates in Archaeological Sites. Journal of Field Archaeology 11: 239-246. Stein, Julie K., 1985, Interpreting Sediments in Cultural Settings. In Archaeological Sediments in Context, p. 5-19, edited by J. K. Stein and W. R. Farrand. Orono: Center for the Study of Early Man. Stein, Julie K., 1991, Scale in Archaeology, Geosciences, and Geoarchaeology. In Effects of Scale on Archaeological and Geoscientific Perspectives, edited by J. K. Stein and A. R. Linse, pp. 1-10. Boulder: Geological Society of America. Stein, Julie K., 2001, Archaeological Sediments in Cultural Environments. In Sediments in Archaeological Context, edited by J. K. Stein and W. R. Farrand, p. 1-28. Salt Lake City: University of Utah Press. Stewart, Iain, 2000, Preface. In The Archaeology of Geological Catastrophes, edited by W. J. McGuire, D. R. Griffiths, P. L. Hancock, and I. S. Stewart, pp. vii-ix. London: The Geological Society. Suess, Hans E., 1955, Radiocarbon concentration in modern wood. Science 122: 415-417. Wallinga, Jakob, Murray, Andrew S., Duller, Geoff A. T., and Tornqvist, Torborn E., 2001, Testing optically stimulated luminescence dating of sand-sized quartz and feldspar from fluvial deposits. Earth and Planetary Science Letters 193: 617-630. Waters, Michael R., 1986, The geoarchaeology of Whitewater Draw, Arizona. Tucson, Ariz.: University of Arizona Anthropological Paper 45. Weiss, Dennis J., 1974, Late Pleistocene stratigraphy and paleoecology of the lower Hudson River estuary. Geological Society of America Bulletin 85: 1561-1570.