Stratigraphic Investigations at

Los Buchillones, a Coastal Taino

Site in North-Central Cuba

Matthew C. Peros,1,* Elizabeth Graham,2 and Anthony M. Davis3

1Laboratory for Paleoclimatology and Climatology, Department of Geography,

University of Ottawa, 60 University Avenue, Ottawa, Ontario, Canada, K1N 6K22Institute of Archaeology, University College London, 31–34 Gordon Square,

London, United Kingdom, WC1H 0PY 3Department of Geography, University of Toronto, 100 St. George Street,

Toronto, Ontario, Canada, M5S 3G3

The authors present stratigraphic data from Los Buchillones, a now submerged Taino villageon the north coast of central Cuba that was occupied from some time prior to A.D. 1220 until1640 or later. Los Buchillones is one of the best-preserved sites in the Caribbean, with mate-rial culture remains that include palm thatch and wooden structural elements from some ofthe more than 40 collapsed structures. The purpose of this study was to investigate the envi-ronment and site-formation processes of the Taino settlement. Sediment cores were sampledfrom the site and its vicinity to permit integration of the geological and archaeological strati-graphies. The cores were analyzed for color, texture, mollusk content, elemental geochem-istry, and mineralogy. The results of the stratigraphic work are consistent with regional sea-level data that shows relative sea level has risen gradually during the late Holocene, but hasremained relatively stable since the time the Taino first occupied Los Buchillones. Of the twostructures partially cleared, at least one appears to have been built over the water, supportedon pilings. Site selection is likely to have resulted from a consideration of environmental fac-tors, such as access to marine, terrestrial, and lagoonal resources, and proximity to freshwa-ter springs. © 2006 Wiley Periodicals, Inc.

INTRODUCTION

Coastlines are complex and dynamic systems that have long been a focus forhuman settlement. The benefits of coastlines for human occupation are numerousand include marine and terrestrial (and, in some cases, estuarine and fluvial)resources within a limited geographic area; suites of organisms unique to the inter-tidal zone, such as crustaceans and edible seaweeds; relatively moderate climateregimes; and direct access to seaborne transportation and travel routes (Bailey andParkington, 1988). Coastal environments are also sensitive to a wide range of mod-ern human activities, which range from urbanization and coastal engineering (Chenet al., 2002) to the exploitation of resources, such as fish and vegetation, especially

Geoarchaeology: An International Journal, Vol. 21, No. 5, 403–428 (2006)© 2006 Wiley Periodicals, Inc.Published online in Wiley Interscience (www.interscience.wiley.com). DOI:10.1002/gea.20113

*Corresponding author; E-mail: mperos@uottawa.ca.

mangroves (Hogarth, 1999, pp. 171–198). It is generally anticipated that coastlines willbe profoundly affected by global warming through relative sea-level rise, whichwould result in flooding, erosion, and saltwater intrusion (Warrick et al., 1993). Giventhe often high population densities associated with coastlines in both the past andthe present, current research into human–environmental interactions in such set-tings (e.g., Schmitt and Osenberg, 1996; NOAA, 2003) is clearly critical.

Contemporary studies of coastal environmental change have been accompaniedby a growing appreciation of the importance of understanding how prehistoric soci-eties adapted to and impacted coastal systems, partly because such studies addgreatly to our knowledge of the chronology of environmental change. Archaeologicalresearch has focused on a number of areas, including subsistence (e.g., Anderson,1988), settlement patterns (e.g., Parkington et al., 1988; Voorhies, 2000), climaticinteraction (e.g., Sandweiss et al., 1996) and coastal-resource exploitation (e.g.,Rippon, 2000). As a result of their location at the interface between marine and ter-restrial environments, coastal archaeological sites generally provide a perspectiveon subsistence, trade, and other aspects of economy that cannot be gleaned from stud-ies focusing exclusively on inland settlements. The study of coastal sites is, how-ever, accompanied by unique difficulties. Many sites are now submerged because ofthe rise in sea level that followed the last deglaciation, tectonic upheaval, and otherless dynamic geomorphological processes (Masters and Flemming, 1983; Johnson andStright, 1992). Furthermore, site preservation is often problematic because waveactivity during the period of submergence has led to the destruction of many near-shore sites (Flemming, 1983).

In the Caribbean, coastal prehistoric sites are numerous (Keegan, 1992a, 1992b;Watters et al., 1992; Watters, 1998; Keegan, 2000; Seidemann, 2001), but compar-atively few have been subjected to intensive geoarchaeological examination. Manyof the coastal sites in the Greater Antilles and the Bahamas were occupied by theTaino, a cultural group which emerged around 800 years B.P. from earlier pot-tery-making farmers who colonized the Caribbean by 2500–2200 yr B.P. Thesepeople and their successors, the Taino and the Island Carib, were horticultural-ists who cultivated corn, manioc, and cotton (Keegan, 2000). Rouse (1992, p. 8)classifies the Taino into three groups: the Western Taino, who occupied theBahamas, most of Cuba, and Jamaica; the “Classic” Taino, who occupied most ofHispaniola and Puerto Rico; and the Eastern Taino who occupied the Virgin andLeeward Islands. The population of Western Taino villages may have been in therange of several hundred occupants (Rouse, 1992, p. 17). In Cuba, the Caribbean’slargest island, estimates of the Taino population range from 150,000 to 1,000,000inhabitants by the 15th century (Dobyns, 1966, p. 409; Dacal Moure and Rivero dela Calle, 1996, p. 20). By the mid-16th century, most of the Taino population underSpanish domination had disappeared or fragmented into small, isolated commu-nities as a result of the combined effects of slavery, warfare, and disease (Rouse,1992, pp. 150–161; Keegan, 2000).

In this article, we have three objectives: (a) to present data on the environmen-tal history and site-formation processes of Los Buchillones, a now submerged Tainovillage occupied from some time prior to A.D. 1220 until 1620–1640 or later

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA404

(Pendergast, 1996a; Pendergast et al., 2002); (b) to discuss the implications of the envi-ronmental history for the understanding of the dynamics of Taino settlement; and (c)to build a database on human–environmental interaction from Los Buchillones thatwill provide significant underpinning for our understanding of long-term Caribbeancoastal adaptations in general.

Dry-land excavation at Los Buchillones began in 1983 under the direction of Dr.Jorge Calvera (Calvera et al., 1996). The reconnaissance stage of the present projectbegan in 1996, with the first excavations in 1997 followed by investigation of portionsof two submerged structures in 1998 and 1999 (Pendergast, 1996b; 1997; Calvera etal., 2001; Pendergast et al., 2001, 2002, 2003). The work has revealed one of the best-preserved prehistoric sites in the Caribbean (Pendergast et al., 2001; see also Conradet al., 2001). To date, investigations have been confined to mapping locations of off-shore standing groups of posts, conducting test excavations in areas of secondarydeposition, exposing but not removing the remains of two structures, and under-taking selected excavations in the terrestrial zone of the site. Because both the struc-tural and the redeposited remains presently lie beneath the sea, excavations haverequired the construction of sandbag and polyethylene sheet dykes around the areasto be investigated, followed by removal of entrapped water with a gasoline-poweredpump. Although continuing inflow of water beneath the dykes has limited the depthof the investigations, work proceeded, in large part, as if Los Buchillones were adry-land site. Our investigation of the stratigraphy involved removal of sedimentcores and their integration with stratigraphic data derived from the archaeologicalexcavations. Here, we present a detailed geoarchaeological study of the LosBuchillones area, as well as new information on Taino settlement patterns that hasresulted from our investigations.

BACKGROUND

Study Area

Los Buchillones (22°22'20"N, 78°48'10"E) is located on the north coast of centralCuba, in the province of Ciego de Avila (Figure 1). A shallow carbonate shelf extendsfar offshore. On its seaward edge lies an active coral reef, the Archipiélago Sabana-Camagüey. The reef is separated from the mainland by the shallow, hypersaline Bahíade Buena Vista. The maximum tidal range is only 30 cm. Despite relatively low-energyconditions, much of the north coast of Ciego de Avila, including parts of the LosBuchillones site, has eroded by over 40 m during the last several decades, probably,in large part, as the result of the construction of a breakwater roughly 3 km east ofthe archaeological site (J. Calvera, personal communication, February, 1997). Thiserosion has been possibly accelerated by the building of a causeway roughly 20 kmto the east of the site that links the mainland to the offshore reef.

Three salt diapirs—named Punta Alegre, Turiguanó, and Cunagua—constitutethe only significant relief in the region (Meyerhoff and Hatten, 1968). The PuntaAlegre and Turiguanó diapirs consist of gypsum overlying halite, limestone, anddolomite (Iturralde-Vinent and Roque Morrero, 1982). Meyerhoff and Hatten (1968)

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

405

report that all three diapirs had pierced the surface by the Pliocene epoch, but thereis no evidence to indicate significant vertical movement of the crust during theHolocene (Iturralde-Vinent and Roque Marrero, 1982).

The Los Buchillones site is situated east of the village of Punta Alegre, mostlywithin the Bahía de Buena Vista and partly within a lagoon (Figure 2). The lagoon ishypersaline and has a maximum length and width of 1.6 km and 300 m, respectively.It is deepest at the eastern end (70 cm) and shallows towards the western end. A che-nier separates the lagoon from the Bahía de Buena Vista.

Hurricanes pass over the Los Buchillones area roughly once every 20 years; themost recent was category-3 Hurricane Michelle in November 2001 (NationalHurricane Center, 2003). Despite the proximity of Hurricane Michelle, observa-tions made in February 2002 indicate that little shoreline erosion occurred at theLos Buchillones site, although lagoons 5 km west of the site were breached some-

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA406

Figure 1. The study area. The Los Buchillones site lies between the towns of Punta Alegre and MaximoGomez on the Bahía de Buena Vista. The elevations of the Punta Alegre, Turiguanó, and Cunagua diapirsare 138, 105, and 315 m, respectively.

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

407

Figure 2. The Punta Alegre lagoon (top) and 2001 excavation area (bottom). In the top figure, transectA�–A was made in 1999; the resulting profile is shown in Figure 7. The thin solid lines on either side of A�–Arepresent the locations of the transects established in 2000. The main concentration of Taino structureslies between the dotted lines. In the bottom figure, the dark grey area represents where excavationsoccurred in 2001. The numbers (1–9, 11, 14, 17, 19–27) designate the excavation squares; the letters (A–E)represent coring sites. A large vertical wooden post (~30 cm diameter) is in square 22. The collapsedstructure excavated in 1999 lies several meters to the northeast of excavation squares 3, 6, and 9.

time between May 2001 and May 2002, possibly as a result of this hurricane. WhileHurricane Lili struck the western end of Cuba in October 2002, local residentsclaim that it had no effect on the northern part of the province of Ciego de Avila.Hurricane Georges was responsible for high winds and flooding in September 1998on the seaward side of the Archipiélago Sabana-Camagüey, although the mainlandexperienced minimal damage.

Archaeology at Los Buchillones

The Los Buchillones site (Pendergast, 1996b; Pendergast et al., 2001, 2002, 2003)covers an area at least 50 m wide and 500 m long, with its long axis obliquely orientedto the chenier (Figure 2). Marine, littoral, lagoonal, and terrestrial environmentsexist at the site. Artifacts have been recovered as far as 1 km west of the area shownin Figure 2, although some of this material may have been redeposited by recenterosion. Offshore submerged structural features extend along the shoreline of theBahía de Buena Vista northeast of the excavation limits shown in Figure 2. The upperboundary of the cultural occupation zone underlies approximately 50 cm of waterand 50–70 cm of sediment. Cultural material recovered from survey and excavationincludes Amerindian stone, pottery, wooden artifacts, faunal and floral remains, andstructural elements of wood and thatch (Graham et al., 2000; Pendergast et al., 2001,2002, 2003). Twenty-three AMS dates on archaeological artifacts and structural tim-bers place the occupation of the site between some time prior to A.D. 1220 and 1640or later (Pendergast, 1996a; Pendergast et al., 2002).

The evidence recovered from survey and limited excavations between 1997and 2001 comprises structural material from, and basic plan data on, more than40 collapsed Taino structures (some of which may be fish weirs or other nonres-idential structures), all of which lie below present sea level. The offshore houselocations are identified by groups of post butts made from lignum vitae (Guaiacum

sp.), palm (Arecaceae), and other tropical hardwoods, possibly including poi-sonwood (Metopium sp.) (Newsom, personal communication, August 9, 2005).Based on the evidence, structures appear to have been of several forms, includ-ing round, rectangular, and possibly oval (Pendergast et al., 2002, p. 79). The struc-ture partially cleared in 1998, which lies partly offshore and partly beneath the pres-ent shoreline (~50 m northeast of the sandbagged zone shown in Figure 2) wasround, whereas the structure uncovered in 1999 in the Punta Alegre lagoon wasrectangular.

The spatial organization of the archaeological remains suggests that relativelylittle lateral postsettlement disturbance has occurred. The remains of the struc-ture cleared in 1999 include rafters and other structural debris, such as thatch, themain elements of which are oriented in a common direction (Figure 3). Althoughno structural elements were removed by the excavators, excavations around andbeneath selected support posts, rafters, and other structural remains revealedthat the sediments burying the elements were water-laid and accumulated underconditions of low energy. The absence of terrestrial sediments and stones trappedbeneath even the heaviest posts indicated that the structures had not collapsed

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA408

on dry land and were then submerged at a later time. Wave activity strong enoughto remove such evidence would also presumably have been strong enough to dis-place the thatch that was encountered in association with the stringers on whichit was originally suspended; many of the collapsed house posts would likely haveshifted as well.

Evidence suggests that the force of impact upon collapse was weak and thatdisturbance to collapsed material was minimal. The structural elements seem tohave been protected from forces such as high winds and rain that would havebeen damaging had the house collapsed on dry land; strong currents or waveactivity would also have been expected to displace or destroy evidence. Our bestexplanation for this level of preservation is that the structure excavated in 1999once stood on pilings in water under protected conditions and collapsed in situ.

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

409

Figure 3. Photograph of the 1999 excavation area inside the Punta Alegre lagoon (view to the north-east). The gasoline pump, mounted on the sandbag dyke, is in operation. In the foreground is structuraldebris from a Taino house; note the common orientation. A Taino house post lies at the base of the ver-tical stick in the center of the image.

Good preservation and lack of disturbance are indicated across all shorelineareas of the site investigated thus far. It, therefore, appears that the village formedby the buildings was built over shallow water; when the settlement was aban-doned, the structures collapsed into a relatively low-energy system that fosteredpreservation. Further archaeological excavation, both extensive and intensive,will be necessary to define fully the depositional matrices of all structures andstructural elements and hence to confirm the shallow-water habitation hypothe-sis. Furthermore, the chemistry of the Los Buchillones lagoonal and near-shore sed-iments will need to be carefully examined because their locally recognized cura-tive properties may indicate that they played a role in preservation of thearchaeological remains.

The results of the research reported here constitute a step towards the clarifica-tion of premodern environmental conditions. Our investigations of the overall sitestratigraphy thus far have allowed us to infer the depositional environments of thesedimentary units associated with the archaeological remains and to determine pat-terns of past sea-level change in the area.

Coastal Geomorphology at Los Buchillones

Kraft and Chrzastowitz (1978) developed stratigraphic models for transgressiveand regressive lagoon-barrier systems. In the transgressive model—which, as willbe shown, is most applicable to the situation at Los Buchillones—the lagoon andbarrier migrate upward and landward in response to a rise in relative sea level. Aslong as the rate of sediment deposition within the lagoon, the rate of relative sea-level rise, and the slope of the coastal plane remain constant, the configuration ofthe lagoon-barrier system will remain unchanged. The resulting stratigraphy is char-acterized by overlapping facies that lie roughly parallel to the coastal plane.

The preservation of coastal facies depends largely on wave energy; wave energyaffects the depth of erosion across the shoreface (Fischer, 1961). High wave energycauses deep shoreface erosion, which on a transgressive coast may result in aloss of earlier offshore sediment. However, sediment composition, shoreline ori-entation, tidal range, topography, and the rate and direction of relative sea-levelchange also influence facies preservation. Belknap and Kraft (1981) also suggestthat a slowly rising sea level will increase the potential for erosion, because waveswill be concentrated along the shoreface for a greater period of time. Conversely,if sea-level rise is rapid, wave action will be temporary, limiting erosion and pos-sibly allowing a greater portion of the offshore stratigraphy to be preserved.

As mentioned earlier, the type of barrier that separates the Punta Alegre lagoonfrom the Bahía de Buena Vista is a chenier. A chenier is a long, narrow, low-lying,shore-parallel ridge, which overlies a muddy substrate and is usually separatedfrom the mainland by a mudflat or lagoon (Otvos and Price, 1979; Waters, 1992,p. 259). Cheniers are geologically ephemeral features, with sizes and positionsinfluenced by factors such as relative sea-level change, sediment availability, andstorm activity (Augustinus, 1989). The chenier at Los Buchillones most likelyformed by the reworking and winnowing of offshore sediment, which produced

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA410

a shell lag that was transported landward and concentrated in a low-lying ridge(Waters, 1992, p. 260). Shells are easily incorporated into cheniers because theirlow specific gravity promotes landward transport into the supratidal zone (Cangziand Walker, 1989). Winnowing may have been an episodic (e.g., during storms) ora continuous process.

METHODS

Field Methods

Geoarchaeological fieldwork at Los Buchillones began in 1999. That year, 11 sed-iment cores (numbered 1 to 11) and two soil pits (1 and 2) were sampled along a ~500-m northwest–southeast trending transect intersecting the main artifact concentra-tion. The sediment cores were recovered with a Livingston piston corer (Wright,1967). The positions of the cores and soil pits were recorded with a Garmin GPS 12.In May 2000, two new transects were established: one 100 m east and the other 150m west of the 1999 transect. The locations of these cores, along with a descriptionof each, were documented in the field. The excavations in May 2001 included areasin the lagoon and in the Bahía de Buena Vista, which were connected by a trench,approximately 1 m deep and 1 m wide, excavated perpendicular to the chenier(Figure 2). The purpose of the trench was to identify the stratigraphy underlying thechenier and to determine whether archaeological material lay beneath it. Five cores(designated A–E) were also sampled from the base of the trench.

Laboratory Methods

The first objective of the study was to generate a shore-perpendicular stratigraphicprofile by identifying and describing the stratigraphic units and correlating themamong adjacent cores. Color and texture were the primary criteria for identifying theunits. Wet colors were recorded with a soil color chart (Oyama and Takehara, 1970).Texture (proportion of sand, silt, and clay) was estimated qualitatively. Subsampleswere then taken from the center of each sedimentary unit in the cores and soils pitsfor more detailed analyses, as described below.

To assist in the correlation of the units among the cores, we attempted to deter-mine if individual stratigraphic units had unique geochemical and/or mineralogicalsignatures. The elemental composition of each subsample was determined on the � 2mm fraction by instrumental neutron activation analysis (INAA). The INAA deter-mined the total concentration of selected major (Al, Ca, Cl, Fe, K, Mg, Na) and trace(As, Ba, Br, Co, Cr, Cs, Ga, Hf, Mn, Ni, Rb, Sb, Sc, Ta, Ti, Th, U, V) elements.

In terms of mineralogy, the relative abundance of primary minerals was deter-mined using a Philips X-ray diffractometer (Philips Electronics, The Netherlands).Samples of bulk sediment were air-dried and disaggregated with a mortar andpestle before analysis. Each sample was X-rayed for 60 minutes. Mineral identi-fication was made using X-pert system software (PANalytical, Almelo, The

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

411

Netherlands) and a searchable powder diffraction file database. The relative abun-dance of the secondary (clay) minerals was determined with a Toshiba ADG-301HX-ray diffractometer (Toshiba, Tokyo, Japan). To separate the clay fraction, thesubsamples were disaggregated and thoroughly mixed in distilled water in hydrom-eter jars. After 24 hours, the suspended clay slurry was siphoned out and cen-trifuged onto ceramic tiles, air-dried, and X-rayed for 60 minutes. The tiles werethen saturated in ethylene glycol vapor at 60°C for 24 hours and X-rayed for 15 min-utes each. Each tile was then heated for 2 hours at 300°C and X-rayed for 15 min-utes. This procedure was repeated at 500°C and 550°C. Interpretation of the dif-fractograms was made by reference to several manuals (Carroll, 1970; Moore andReynolds, 1989).

The proportion of organic matter, carbonates, and silicates in each subsamplewas estimated by loss on ignition, following Dean (1974). Our assumption was thatunits with relatively high silicate content were more likely to be of terrestrial origin,whereas high carbonate values would likely represent shell material and be indica-tive of sediments deposited in marine or lagoonal settings.

To determine molluscan ecology, mollusks were identified to genus or specieslevel in cores 3 and 5. Sediment samples were wet-sieved through a 63-µm stain-less steel sieve and all identifiable mollusks were retained and air-dried. Referencemanuals were used for identification (Warmke and Tucker Abbott, 1961; TuckerAbbott, 1974).

RESULTS

Stratigraphy

Based on similarities in color and texture, six sedimentary units were identi-fied (Table I), and the resulting stratigraphic profile is illustrated in Figure 4. Atthe base of the section is unit 1, a silty to pebbly unit that extends at least 300 moffshore and continues onshore south of the lagoon. The cores sampled in May2000 confirm that this red-brown unit also extends east and west of the 1999 tran-sect, at roughly the same slope and depth below mean sea level. Overlying unit 1is unit 2, a grayish-brown silty-clay that underlies the back-barrier. A lens of fine,light-grey sediment lies within unit 2 but has not been assigned to a separate sed-

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA412

Table I. Unit descriptions.

Unit Color Texture Description

1 Red-brown (7.5Y 4/4) Silty to pebbly Stiff and dry; shell and roots rare2 Grey-brown (7.5 Y 4/2) Silty-clay Moist and slightly plastic, no shell or roots3 Olive-gray (5Y 5/1) Clayey No shell; poorly preserved turtle grass

(Thalassia testudinum) roots are common 4 Light-gray (5Y 4/3) Silty-clay No shell or roots5 Light-gray (5Y 6/1) Silty-clay Abundant shells and roots6 Olive-gray (5Y 5/1) Sandy Abundant shells and shell fragments

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

413

Fig

ure 4

.St

rati

grap

hy a

long

tra

nsec

t A

�–A

. Cor

es a

nd s

oil p

its

are

repr

esen

ted

by t

he v

erti

cal b

lack

bar

s. A

rcha

eolo

gica

l mat

eria

l is

conc

entr

ated

wit

hin

the

dash

ed a

rea,

alt

houg

h ex

tend

s do

wnw

ard

into

uni

t 3.

The

ver

tica

l exa

gger

atio

n is

64�

.

imentary unit at this time. Unit 3 is clayey and olive-grey. It lies offshore and doesnot extend into the present lagoon. Unit 4 underlies the back-barrier and has a silty-clay texture with yellow-mottles. Unit 5, also consisting of silty-clay, is directlyassociated with the Punta Alegre lagoon. Unit 6, dominated by shell fragmentsand turtle grass (Thalassia testudinum) roots, represents the most recent marinedeposit. The lateral relationship between units 5 and 6 is unclear, but the twounits seem to grade into each other beneath the chenier. Based on the overlappingpositions of the facies, the relative age of these units from oldest to youngest isunit 1, unit 2, unit 3, unit 4, unit 5, and unit 6.

The chenier consists of well-sorted mollusks, in particular Cerithium eburneum

and Chione cancellata, with smaller amounts of Modulus modulus, Polinices lac-

teus, and Anomalocardia auberiana. The shells of the chenier grade into shelly mudat ~30 cm below the surface that buries wooden Taino structural material located ~70cm below the surface of the chenier.

Elemental Geochemistry

The concentrations of select major and trace elements for each sample are shownin Table II. A correspondence analysis was performed on these data to display therelationship between the subsamples in a two-dimensional scattergram (Figure 5).Individual subsamples are plotted by sedimentary unit membership. The samplesform distinct clusters, indicating that each unit has a relatively homogenous and dis-tinctive elemental composition.

When the subsamples are seriated by unit membership along the first dimensionof the correspondence analysis (Baxter, 1994, pp. 118–123), the ordering of theunits is identical to their relative chronology. This indicates a progressive changein the elemental composition of successive facies, with the two end-members ofthis sequence—units 1 and 6—representing the most dissimilar environments (ter-restrial and marine, respectively). The positions of the intermediate units—2, 3, 4,and 5—indicate that the marine environment has had an increasingly importantinfluence on the compositions of the facies. The reason for the concordancebetween relative chronology and elemental composition can be explained withreference to the pronounced arch evident among the sample points. This archeffect often occurs when samples are measured along an environmental gradient(Legendre and Legendre, 1998, p. 465); in this case, the gradient is elemental com-position over time.

Mineralogy

Results from the mineralogical analysis of the � 2 mm fraction show that quartzand calcite, with some gypsum, constitute the bulk of the material in each unit, withroughly equal abundances (Peros, unpublished data). The presence of gypsum isnot surprising given the proximity of the Punta Alegre diapir. Individual units aredistinctive, however, when the mineralogy of the clay-sized (� 2 µm) fraction is con-sidered (Table III). Unit 6 is distinguished by the presence of smectite and high illite-

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA414

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

415

Tab

le I

I. C

once

ntra

tion

s of

maj

or a

nd t

race

ele

men

ts.

Uni

tSa

mpl

eA

l%C

a%C

l%F

e%K

%M

g%N

a%A

s B

a B

r C

o C

r C

s H

f M

n R

b Sc

Ta

T

i T

h U

V

11

4.35

8.66

0.77

3.81

1.47

1.97

0.24

1.98

150

3329

.315

11.

783.

8374

151

.119

.11.

4065

833.

332.

0214

72

4.92

7.55

0.99

4.66

1.55

1.89

0.35

4.68

169

3852

.115

21.

483.

9390

746

.719

.41.

4571

953.

171.

9615

73

4.45

4.79

1.52

5.16

1.65

1.93

0.10

2.65

177

5831

.517

11.

994.

5563

758

.622

.01.

4174

513.

562.

7615

24

4.55

9.08

0.93

4.70

1.42

1.83

0.29

2.41

152

4523

.216

91.

593.

9349

741

.219

.21.

5264

553.

112.

0514

55

3.98

6.91

1.03

4.74

1.49

1.99

0.12

2.76

154

4223

.118

22.

003.

9779

554

.018

.81.

4159

833.

051.

5314

86

4.10

8.59

0.79

4.20

1.27

1.56

0.22

3.18

155

3520

.613

21.

263.

5637

141

.616

.41.

4060

542.

561.

5214

37

3.30

12.3

00.

685.

011.

091.

370.

194.

2318

929

25.4

911.

132.

9330

535

.614

.81.

1946

512.

151.

5213

38

5.86

8.15

0.68

3.27

1.32

1.20

0.22

3.74

130

2728

.510

61.

083.

0710

400

35.3

13.7

1.15

�41

532.

191.

3915

79

4.23

6.62

0.81

3.27

1.22

�1.

200.

121.

9611

238

14.8

143

1.10

3.62

361

35.7

14.7

1.24

6463

2.23

1.59

122

104.

007.

180.

123.

741.

091.

070.

142.

1210

919

16.3

123

1.26

3.24

411

41.1

15.8

1.17

4899

2.41

1.78

103

113.

9414

.30

0.03

3.47

0.94

1.25

0.06

2.59

111

3115

.414

31.

122.

8436

034

.614

.50.

9650

372.

271.

3211

5

212

5.09

8.13

1.99

4.16

1.90

2.77

0.24

�1.

4017

689

23.4

162

1.53

3.33

1188

53.3

18.4

1.16

5602

2.99

3.68

153

134.

338.

611.

883.

871.

882.

280.

15�

1.20

122

7725

.616

21.

683.

0769

956

.017

.41.

0953

852.

736.

4513

514

4.92

9.37

1.76

4.44

2.18

2.78

0.23

�1.

4012

074

24.7

156

1.95

3.61

612

65.0

19.6

1.19

5429

3.15

4.98

156

154.

837.

861.

544.

711.

772.

330.

103.

9615

763

27.2

160

1.70

3.42

617

53.3

18.2

1.21

5829

2.94

�0.

8013

516

3.34

5.08

0.85

3.51

1.33

1.54

0.19

1.54

114

3618

.520

91.

424.

0377

743

.514

.01.

1658

012.

311.

6611

617

3.67

7.77

0.95

3.74

1.26

1.62

0.12

3.85

113

3926

.814

81.

453.

8132

844

.616

.41.

2157

922.

622.

0313

3

318

4.18

10.1

00.

844.

611.

421.

590.

253.

3613

337

24.8

151

1.67

2.65

917

44.2

17.7

1.07

5894

2.94

2.61

144

193.

597.

791.

183.

671.

311.

570.

092.

3911

146

18.6

147

1.73

2.59

1061

45.9

18.3

1.09

6283

3.05

2.21

115

204.

578.

420.

954.

401.

351.

630.

292.

7315

639

23.2

152

1.21

2.49

1157

46.6

15.5

1.22

5853

2.42

3.68

153

421

3.60

10.5

01.

883.

581.

541.

920.

212.

3012

684

19.2

194

1.57

2.86

977

45.3

15.3

0.98

4280

2.24

3.72

119

222.

9313

.10

2.39

3.81

1.12

1.46

0.17

4.72

9095

15.8

286

1.24

1.72

648

31.1

11.8

0.64

3592

1.34

1.90

9923

4.32

8.50

1.63

4.40

1.81

2.40

0.22

6.28

9270

24.5

174

1.52

3.18

814

50.6

16.5

1.13

4741

2.51

1.65

150

524

3.40

13.3

02.

453.

671.

501.

510.

113.

17�

4595

36.4

305

1.41

1.97

1549

51.0

15.5

0.79

4457

1.79

4.47

6925

2.29

16.4

01.

712.

511.

223.

190.

05�

0.90

8072

15.2

167

1.16

1.84

1081

37.9

11.0

0.65

2975

1.54

3.70

7526

2.55

19.0

01.

592.

401.

152.

840.

254.

8563

6815

.513

61.

201.

6569

337

.210

.20.

6525

351.

423.

8565

272.

6311

.00

1.72

3.04

1.34

3.48

0.03

5.81

9364

20.0

167

1.47

2.32

844

42.1

12.6

0.80

3036

1.79

2.56

100

282.

2818

.30

1.09

2.52

1.01

2.21

0.21

9.30

�34

4216

.013

40.

741.

7713

8028

.28.

940.

6030

741.

243.

1572

291.

7417

.80

1.24

1.67

0.62

1.41

0.08

3.47

�28

4510

.715

30.

831.

6056

623

.17.

690.

5025

980.

983.

0446

302.

1614

.10

1.18

2.05

0.80

1.26

0.17

2.51

7560

16.7

189

1.25

3.26

517

31.3

10.5

0.71

4006

1.52

2.71

88

631

2.23

18.3

02.

241.

960.

690.

680.

323.

4969

112

14.2

243

0.83

1.24

246

18.5

8.46

0.36

�76

90.

808.

6447

321.

0016

.80

1.15

0.70

0.28

0.44

0.05

1.12

5145

5.6

220

�0.

201.

8626

46.

33.

770.

2714

510.

543.

7926

smectite, whereas unit 3 contains relatively high quantities of illite, chlorite, illite-smec-tite, and kaolinite. Units 1, 2, 4, and 5 have similar clay-mineral compositions; thisfinding suggests that many of the materials that constitute these units are derived fromthe same terrestrial source. The smectite in unit 6 may represent an authigenic marineclay, such as glaucony (Weaver, 1989, p. 372), although more detailed analysis isrequired to confirm this hypothesis. Thus, the results show that unit 6 has a uniquemineralogy when the � 2-µm fraction is considered, although the other units aredifficult to differentiate on the basis of mineralogy alone.

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA416

Figure 5. Correspondence analysis (top) and seriation (bottom) of the elemental data in Table III. Thedashed arrow shows how samples are seriated from the correspondence analysis.

Organic Matter, Carbonates, and Silicates

Loss-on-ignition results for two representative cores are shown in Figure 6.Among all cores, unit 3 contains the most organic matter at ~12%, whereas unit 1has the least at ~6%. In all six units, carbonate constitutes 20–40% of the sedi-ment and generally decreases in abundance with depth, whereas silicates con-stitute the bulk of each core (� 60%) and generally increase in abundance withdepth. These results are consistent with the visual descriptions of each core; theunits dominated by shell (5 and 6) contain the highest carbonate values, whereasunit 1, with almost no shell, yields the highest amount of silicate, consistent withits classification as a terrestrial unit.

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

417

Table III. Mineralogy of clay-sized fraction.

Clay Minerals Primary Minerals

Unit Sample Smectite Illite Chlorite Illite-Smectite Kaolinite Calcite Gypsum Quartz Orthoclase

1 1 — � � � — � � � —2 — — � — — � � � —3 — � �� � � � � � —4 — �� �� �� �� — � �� —5 — �� �� � �� �� � �� ��

6 — �� �� �� �� � � � �

7 — — — � — — — � —8 — — �� � — � — � —9 — � �� � — � � � —

10 — � �� �� � � � � —11 — � � — — �� — � —

2 12 — � � � � �� � � —13 — �� �� � �� �� � �� —14 — � � � �� �� — � —

3 17 — ��� �� �� �� � � �� ��

18 — �� �� �� �� � � � �

19 — ��� ��� �� ��� � � �� ��

4 21 — � �� � � � � � —22 — — � — — �� � � —23 — � � � — � � � —

5 24 — — � � � �� � � —25 — — — — � � �� — —26 — — � — � — �� � —27 — — � � � � �� � �

28 — �� �� �� — — � � —29 — � �� � �� � — � �

30 — � �� � — � — � —

6 31 �� �� — �� � � � �� �

32 �� �� �� �� � �� — � �

Note. ��� � abundant; �� � moderate; � � trace; — � not present.

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA418

Figure 6a. Loss on ignition and mollusk analysis results. The loss-on-ignition data are expressed as a per-centage of the total dry weight of the sample. The mollusk results are total number of mollusks countedper taxon for each sample interval. Core 3 (nearshore).

Molluscan Ecology

The total number of each identified mollusk was plotted against the stratigraphyfor cores 3 (Figure 6a) and 5 (Figure 6b). The bivalves (Anomalocardia, Chione,and most of the unknown examples) almost always exist in halves, and to maintainwhole numbers, each half was counted individually, with the result that the bivalvesare overrepresented. Mollusks are associated only with units 5 and 6. In order ofabundance, they are C. eburneum, A. auberiana, M. modulus, C. cancellata,Marginella apicina, and Urosalpix spp. All are common in shallow marine condi-tions in the northern Caribbean (Warmke and Tucker Abbott, 1961; Tucker Abbott,

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

419

Figure 6b. Loss on ignition and mollusk analysis results. The loss-on-ignition data are expressed as a per-centage of the total dry weight of the sample. The mollusk results are total number of mollusks countedper taxon for each sample interval. Core 5 (Punta Alegre lagoon).

1974). The proportions of each taxon are consistent down each core. Although onlycores 3 and 5 were studied, a visual examination of the other cores indicates that pat-terns found in these units are consistent throughout the profile.

DISCUSSION

Environments of Deposition

Unit 1 is a buried soil. This is evidenced by its red-brown oxidized appearance,its relatively coarse texture, its lateral continuation onto the present-day surface,and its relatively high silicate content. Unit 1 predates the Taino settlement anddoes not appear to have been an occupation surface. In terms of soil classifica-tion, this unit has physical properties consistent with the Ferralsol group of theFood and Agriculture Organization (Bridges, 1997, pp. 128–130). It is also similarto the Cuban Ministry of Agriculture’s Francisco series soil, which is brownish-redand fine textured with occasional coarse particles (Suelos de la Provincia Ciegode Avila, 1983, p. 127).

The fine texture of units 2, 3, 4 and 5, and their grayish color, indicative of reducedconditions, suggest these facies were deposited in a lagoon. The yellow mottles ofunit 4 may have been caused by alternating reducing and oxidizing conditions froma period of fluctuating or low water levels (Birkeland, 1999, p. 134). The marine mol-lusk fragments in unit 5 indicate that there has been a nearly constant exchange ofwater between the Bahía de Buena Vista and the Punta Alegre lagoon. Despite thisfinding, thick mollusk layers, which would indicate washover fans from hurricanestrikes (Liu and Fearn, 2000), are absent from the cores. This absence of thick mol-lusk layers suggests that hurricanes had a limited impact as a geomorphologic agent,possibly because the Archipiélago Sabana-Camagüey reef protects the mainlandfrom storm surges. Until more cores are studied, however, the presumption of lim-ited hurricane impact will remain tentative.

Unit 6 was deposited in a shallow marine environment. This is indicated by thepresence of turtle grass (Thalassia testudinum), usually in growth position and ahigh abundance of marine mollusks. The coarseness of the sediment in this unit isconsistent with the higher energy conditions of the Bahía de Buena Vista comparedto the Punta Alegre lagoon.

Relative Sea-Level Change at Los Buchillones

The overlapping positions of the sedimentary units, as shown in Figure 4, can beinterpreted using principles of sequence stratigraphy (Van Wagoner et al., 1988).Sequence stratigraphy uses the boundaries that separate sedimentary units to inferthe direction and rate of relative sea-level change (Haq, 1991). The main unit ofanalysis in sequence stratigraphy is the sequence, which is a stratigraphic unitbounded at its top and base by unconformities (Van Wagoner et al., 1988). Suchunconformities, or sequence boundaries, form during periods of low sea level whenthe shallow marine environment is aerially exposed and is subject to erosion. The

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA420

sediments that make up the sequence are deposited during a period of high sealevel occurring between two successive periods of low sea level.

Figure 4 represents the lower portion of a sequence. The contact separating unit1, the buried soil, from units 2, 3, and 5 represents a basal sequence boundary and,therefore, a period of relative sea-level rise. There is no upper sequence boundary,because relative sea level has yet to fall, although the mottled, oxidized appearanceof unit 4 may reflect a brief stillstand. The rise in relative sea level inferred from thestratigraphic sequence is strongly supported by the geochemical results (Figure 5),which indicate that the elemental compositions of lagoonal units 3, 4, and 5 areincreasingly similar to the marine facies (unit 6)— a condition that would be expectedduring a sea-level transgression. A consequence of the relative sea-level rise wouldhave been the landward migration of the lagoon, which can indeed be tracked in theprofile by examining the overlapping positions of units 3, 4, and 5. This migration mayhave included a concurrent landward shift of the chenier, which is a typical responseof such features during a rise in relative sea level. The presence of the chenier over-lying Taino material culture indicates that the chenier reached its present locationsometime since that portion of the site was last occupied.

The units in each core (with the exception of unit 1) show a sequence of finersediment overlain by coarser sediment. This textural pattern represents a shallow-ing-upward sequence, indicating that until the recent period of erosion, sedimentsupply exceeded the rate of relative sea level rise in this area (Nichols, 1989), andhence the lagoon was deeper when it was farther seaward and that it has been grad-ually infilling as it migrated landward. In addition, the orientation of the shorelineappears to have changed as this migration occurred. Assuming the long axis of theLos Buchillones site was parallel to the shoreline (as was the case with many Tainosites in the Bahamas; Keegan, 1992a), the coastline had a slightly more north–southorientation when the site was occupied.

Figure 7 shows several published relative sea-level records from Florida and theBahamas (Scholl et al., 1969; Boardman et al., 1988; Gelsanliter, 1996). These data—obtained from 14C-dated mangrove peat—indicate that relative sea level in the northernCaribbean region was around 2 m below present level at 3000 14C yr B.P. and then roseat a decelerating rate to its present position—a pattern also consistent with recentHolocene sea-level observations from the Mississippi Delta (Törnqvist et al., 2004). Otherresearch from Florida, however, based on geoarchaeological (Walker et al., 1994, 1995)and rhyzolith evidence (Froede, 2002), suggests that centennial-scale highstands of ~1m occurred from 1000 to 2000 14C yr B.P. There is no evidence, such as raised beachesor wave-cut notches, for a Holocene highstand in the Los Buchillones area. Given the tec-tonic stability of the region (Iturralde Vinent and Roque Marrero, 1982), the stratigra-phy from Los Buchillones appears to be largely consistent with the overall pattern andtiming of relative sea-level rise shown in the sea-level curves in Figure 7.

Site Formation and Implications for the Taino Settlement

These investigations provide insight into the environmental setting of the Tainovillage. If relative sea level was 0.2–0.5 m lower during the Taino occupation of Los

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

421

Buchillones (Figure 7), most of the features represented by the archaeological mate-rial, which lie at least 1 m below present sea level, would have been associated withshallow-water conditions. The full range of environmental conditions contemporarywith site formation are, at present, difficult to determine, although a lagoonal ornearshore system is likely. However, given the size of the village and the duration overwhich it was occupied, various components of the site may have been built in differentsettings; some houses, for example, might have been built on a chenier, whereasothers may have been built in the lagoon. Alternatively, the site may have been ini-tially settled on a dry surface, and as the population grew, or relative sea level rose,the Taino adapted by expanding the settlement over water. Nevertheless, the littoralzone at the Los Buchillones site appears to have had a complex history, and it islikely that the Taino were compelled to adapt their settlement patterns to an ever-changing coastal environment.

The vertical wood pilings are consistent with the idea that the Taino houseswere raised above water. Pile dwellings are not uncommon archaeologically. TheLagunillas Phase (3000–1700 yr B.P.) for the Lake Maracaibo Basin in Venezuela ischaracterized by pile dwellings (Wagner and Tarble de Ruíz, 1975) and numerouspile dwellings have been documented from lakes in Germany and Switzerland

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA422

Figure 7. Relative sea-level curves for northern Caribbean region: (1) Scholl et al. (1969); (2) Boardmanet al. (1988); and (3) Gelsanliter (1996). The approximate duration of Taino occupation at Los Buchillonesis represented by the grey bar.

(Hafner, 1995; Strobel, 2000). Indeed, modern pile dwellings can also be found inSoutheast Asia and northern South America. In Belize, small structures are builtover water to accommodate people during periods of intensive fishing activity.The motives for living over water in pile dwellings vary, but include efficient trans-portation and waste-disposal, coping with occasional high tides and flooding, adapt-ing to long-term environmental change (e.g., sea- or lake-level change), ensuringstructural stability on soft substrates, and architectural preference (Oliver, 1987).At Los Buchillones, ensuring structural stability in the mud may have been impor-tant in addition to coping with the wet substrate. Despite this, it is important to notethat raised floors are a necessary requisite for raised houses, and, as yet, floors havenot been identified archaeologically (Pendergast et al., 2001). In any event, piledwellings would represent a unique adaptation among Taino settlements, as coastalTaino sites elsewhere are normally associated with sand beaches (Sears andSullivan, 1978; Keegan, 1992a, 1992b).

Based on the results and discussion above, we have developed a preliminarymodel for the evolution of the coastline (Figure 8). During the time of Taino occu-pation (~A.D. 1220–1640), the Punta Alegre lagoon and chenier were seaward oftheir present position, with much, or perhaps all, of the Taino settlement locatedwithin the lagoon. By the mid-18th century, the village had been abandoned for atleast 100 years and the structures had collapsed into the low-energy environmentof the lagoon. The chenier began to transgress the site, changing its orientationand reaching its present position by the early 20th century. The western, shal-lower end of the lagoon was also inundated gradually during this time. Beginningin the 1950s, erosion of the coast had begun, and the chenier narrowed to its pres-ent width.

On a broader level, environmental considerations appear to have been critical forsite selection. Nearby springs in the Punta Alegre diapir (Figure 2; Fagundo et al.,1993) would have provided potable water, as the next nearest source is the Rio LosPerros 7 km to the west. In addition, the summit of the diapir provides an excellentview of the region. The vast diversity of materials found in the excavations atteststo the exploitation of a variety of micro-environments. Resources would have beenimmediately available from terrestrial, marine, and lagoonal environments, whereasestuarine (the Rio Los Perros estuary) and reef communities could be accessed.Furthermore, the location of the site inside the Archipiélago Sabana-Camagüeyappears to have provided some protection from hurricane activity. Last, the lateoccupation of Los Buchillones (~A.D. 1640) is atypical (Rouse, 1992, p. 158). ThePunta Alegre diapir and Archipiélago Sabana-Camagüey may have been sufficientphysical barriers to hide the site from European contact until the first half of the17th century (Pendergast et al., 2001).

CONCLUSIONS

Geoarchaeological research at Los Buchillones has revealed valuable informa-tion about the complex history of the coastal environment. The stratigraphy showsthat relative sea level has gradually risen over the past several thousand years, and

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

423

there is no evidence to indicate that it has been above its present level during theHolocene. The rise in sea level appears to have been accompanied by a landward shiftin the Punta Alegre lagoon and chenier. The geological and archaeological evidencesuggests that the Los Buchillones site was built above a wet substrate—a settlementstrategy unique among Taino sites elsewhere. Despite this, there is still a lack ofimportant evidence, such as house floors, which would be expected in such cir-cumstances, although future archaeological excavations will shed light on this.Furthermore, given the length of Taino occupation of Los Buchillones, it is notunlikely that the local environment—and consequently Taino settlement patterns—changed during the roughly 400 years the site was occupied.

Ongoing and future work is focusing on several areas. At the site itself, deeperexcavations over a larger area are essential to fully reconstruct Taino house form.

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA424

Figure 8. Preliminary model of the evolution of the Punta Alegre lagoon since the occupation of the site.The fine dashed line represents the present shoreline of the lagoon. The coarse dashed lines are the outerboundaries of the main artifact concentration. Transect A�–A (Figure 7) is also shown.

Our intention is to drive semipermanent caissons into the sediment to allow excava-tions to reach archaeologically sterile sediment without flooding. In addition to pro-viding valuable archaeological information, this method will allow us to examine thesediment directly underlying and overlying the houses to provide a more detailed pic-ture of the depositional environments prior to, during, and following settlement. Oneof us (Peros) is also examining foraminiferal assemblages from the sediments at thesite to determine whether lagoonal and marine facies can be easily identified on thebasis of micropaleontological content. On a regional level, archaeological surveyshave begun on the cays north of Punta Alegre and several small sites with ceramicsidentical to those at Los Buchillones have been found. We have taken sediment coresfrom mangrove swamps on many of these cays and on the mainland west of PuntaAlegre to develop a well-dated relative sea level curve for the area and to reconstructpast vegetation communities. This work will better constrain the timing of inundationof the Los Buchillones area and provide a regional environmental framework withinwhich the archaeology at Los Buchillones and its environs can be interpreted.

ACKNOWLEDGMENTS

Financial support for this study was gratefully received from the Los BuchillonesProject (officially the Proyecto André Bekerman) and from the Department ofGeography, University of Toronto. We thank Dr. Jorge Calvera, Dr. Juan Jardines,and Dr. David Pendergast for permission to work at Los Buchillones. Dr. Bill Mahaneyand Dr. Ron Hancock gave valuable comments and permitted samples to be ana-lyzed in their labs. Dr. Andrew Stewart, Dr. David Smith, and Michael Milner pro-vided valuable discussion. Fieldwork could not have been completed without thehelp of Adrián Garcia L., Pedro Guerra H., Dr José Herrera A., José La O, and NelsonRodríguez T. Sarah Copland provided valuable editorial advice.

REFERENCES

Anderson, A.J. (1988). Coastal subsistence economies in prehistoric southern New Zealand. In G. Bailey& J. Parkington (Eds.), The archaeology of prehistoric coastlines (pp. 93–101). Cambridge: CambridgeUniversity Press.

Augustinus, P.G.E.F. (1989). Cheniers and chenier plains: A general introduction. Marine Geology, 90,219–229.

Bailey, G., & Parkington, J. (1988). The archaeology of prehistoric coastlines: An introduction. In G. Bailey& J. Parkington (Eds.), The archaeology of prehistoric coastlines (pp. 1–10). Cambridge: CambridgeUniversity Press.

Baxter, M.J. (1994). Exploratory multivariate analysis in archaeology. Edinburgh: Edinburgh University Press.Belknap, D.F., & Kraft, J.C. (1981). Preservation potential of transgressive coastal lithosomes on the U.S.

Atlantic shelf. Marine Geology, 42, 429–442.Birkeland, P.W. (1999). Soils and geomorphology. New York: Oxford University Press.Boardman, M.R., Neumann, A.C., & Rasmussen, K.A. (1988). Holocene sea level in the Bahamas. In J.E.

Mylroie (Ed.), Proceedings of the 4th Symposium on the Geology of the Bahamas (pp. 45–52). Bahamas:Bahamian Field Station.

Bridges, E.M. (1997). World soils. New York: Cambridge University Press.Calvera R.J., Serrano, E., Rey, M., Pedroso, I., & Yparraguierre, Y. (1996). El sitio arqueológico Los

Buchillones [The Los Buchillones Archaeological Site]. El Caribe Arqueológico, 1, 59–67.

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

425

Calvera R.J., Pendergast, D.M., Brito M.O., Bekerman, A., Jardines M.J.E., et al. (2001). Investigacionesarqueológicas conjuntas cubano-canadienses en el area de Los Buchillones, Informe Final [Cuban-Canadian Archaeological Investigations in the Los Buchillones Area, Final Report]. Unpublished finalreport. Ciégo de Avila, Cuba: Centro de Ciencia, Tecnología y Medio Ambiente.

Cangzi, L., & Walker, H.J. (1989). Sedimentary characteristics of cheniers and the formation of chenier plainsof East China. Journal of Coastal Research, 5, 353–368.

Carroll, D. (1970). Clay minerals: A guide to their x-ray identification. Geological Society of AmericaSpecial Paper 126. Boulder, CO: Geological Society of America.

Chen, J., Eisma, D., Hotta, K., & Walker, H.J. (Eds.). (2002). Engineered coasts. Dordrecht: Kluwer. Conrad, G.W., Foster, J.W., & Beeker, C.D. (2001). Organic artifacts from the Manantiel de la Aleta,

Dominican Republic: Preliminary observations and interpretations. Journal of Caribbean Archaeology,2, 1–20.

Dacal Moure, R., & Rivero de la Calle, M. (1996). Art and archaeology of pre-Columbian Cuba. Pittsburgh,PA: University of Pittsburgh Press.

Dean, W.E. (1974). Determination of carbonate and organic matter in calcareous sediments and sedi-mentary rocks by loss on ignition: Comparison with other methods. Journal of Sedimentary Petrology,44, 242–248.

Dobyns, H.F. (1966). An appraisal of techniques with a new hemisphere estimate. Current Anthropology,7, 395–416.

Fagundo, J.R., Rodriguez, J.E., de la Torre, J., Arencibia, J.A., & Forti, P. (1993). Hydrologic and hydro-chemical characterization of the Punta Alegre gypsum karst (CUBA). In A. Afrasiabian (Ed.),International Symposium on Water Resources in Karst with a Special Emphasis on Arid and SemiArid Zones (pp. 485–498). Islamic Republic of Iran: Shiraz.

Fischer, A.G. (1961). Stratigraphic record of transgressing seas in light of sedimentation on Atlantic Coastof New Jersey. The American Association of Petroleum Geologists, 45, 1656–1666.

Flemming, N.C. (1983). Survival of submerged Lithic and Bronze Age artifact sites: A review of case his-tories. In P.M. Masters & N.C. Flemming (Eds.), Quaternary coastlines and marine archaeology: Towardsthe prehistory of land bridges and continental shelves (pp. 135–174). London: Academic Press.

Froede, C.R. (2002). Rhizolith evidence in support of a late Holocene sea-level highstand at least 0.5 mhigher than present at Key Biscayne, Florida. Geology, 30, 203–206.

Gelsanliter, S. (1996). Holocene stratigraphy of the Chatham River region, southwest Florida with areevaluation of the Late Holocene sea-level curve. Unpublished master’s thesis, University of Miami,Miami, FL.

Graham, E., Pendergast, D.M., Calvera R.J., & Jardines M.J. (2000). Excavations at Los Buchillones, Cuba.Antiquity, 74, 263–264.

Hafner, A. (1995). Die frühe Bronzezeit in der Westschweiz: Funde und Befunde aus Siedlungen, Gräbernund Horten der entwickelten Frühbronzezeit [The early bronze age in Western Switzerland: Finds andinterpretations from settlements, burial sites and caches of the developed early bronze age]. Bern:Staatlicher Lehrmittelverlag.

Haq, B.U. (1991). Sequence stratigraphy, sea-level change, and significance for the deep sea. SpecialPublications of the International Association of Sedimentology, 12, 3–39.

Hogarth, P.J. (1999). The biology of mangroves. Oxford: Oxford University Press.Iturralde-Vinent, M.A., & Roque Marrero, F.D. (1982). Nuevos datos sobre las estructuras diapiricas de Punta

Alegre y Turiguano, en la Provincia Ciego de Avila [New data about the diapiric structures of PuntaAlegre and Turiguano in the Province of Ciego de Avila]. Ciencias de la Tierra y del Espacio, 4, 47–55.

Johnson, L.L., & Stright, M. (Eds.). (1992). Paleoshorelines and prehistory: An investigation of method.Boca Raton, FL: CRC Press.

Keegan, W.F. (1992a). The people who discovered Columbus: The prehistory of the Bahamas. Gainesville,FL: University Press of Florida.

Keegan, W.F. (1992b). Lucayan settlement patterns and recent coastal changes in the Bahamas. In L.L.Johnson & M. Stright (Eds.), Paleoshorelines and prehistory: An investigation of method (pp. 5–14).Boca Raton, FL: CRC Press.

Keegan, W.F. (2000). West Indian archaeology. Vol. 3: Ceramic age. Journal of Archaeological Research,8, 135–167.

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA426

Kraft, J.C., & Chrzastowski, M.J. (1978). Coastal stratigraphic sequences. In R.A. Davis (Ed.), Coastal sed-imentary environments (pp. 625–663). New York: Springer-Verlag.

Legendre, P., & Legendre, L. (1998). Numerical ecology. Amsterdam: Elsevier.Liu, K.-B., & Fearn, M.L. (2000). Reconstruction of prehistoric landfall frequencies of catastrophic hurri-

canes in Northwestern Florida from lake sediment records. Quaternary Research, 54, 238–245.Masters, P.M., & Flemming, N.C. (Eds.). (1983). Quaternary coastlines and marine archaeology: Towards

the prehistory of land bridges and continental shelves. London: Academic Press.Meyerhoff, A.A., & Hatten, C.W. (1968). Diapiric structures in Central Cuba. American Association of

Petroleum Geologists Memoir, 8, 315–357.Moore, D.M., & Reynolds, R.C. (1989). X-ray diffraction and the identification and analysis of clay min-

erals. Oxford: Oxford University Press.National Hurricane Center. (2003). Seasons Archive. Retrieved November 30, 2003, from http://

www.nhc.noaa.gov Nichols, M.M. (1989). Sediment accumulation rates and relative sea-level rise in lagoons. Marine Geology,

88, 201–219.NOAA. (2003). Coastal zone management. Retrieved June 16, 2004, from http://www.ocrm.nos.noaa.gov/czm/ Oliver, P. (1987). Dwellings: The house across the world. London: William Stout.Otvos, E.G., & Price, A. (1979). Problems of chenier genesis and terminology—An overview. Marine

Geology, 31, 251–263.Oyama, M., & Takehara, H. (1970). Revised standard soil colour charts. Tokyo: Japan Research Council

of Agriculture, Forestry and Fisheries.Parkington, J., Poggenpoel, C., Buchanan, B., Robey, T., Manhire, T., & Sealy, J. (1988). Holocene coastal

settlement patterns in the western Cape. In G. Bailey & J. Parkington (Eds.), The archaeology of pre-historic coastlines (pp. 22–41). Cambridge: Cambridge University Press.

Pendergast, D.M. (1996a). AMS dates from Los Buchillones, Cuba. NewsWARP, 20, 33.Pendergast, D.M. (1996b). The Los Buchillones site, North Coastal Cuba. NewsWARP, 19, 3–6.Pendergast, D.M. (1997). Up from the shallows: A look at the ROM’s first archaeological excavation in Cuba.

Rotunda, 30, 28–35.Pendergast, D.M., Calvera R.J., Jardines, M.J., Graham, E., & Brito, M.O. (2003). Construcciones de Madera

en el Mar, Los Buchillones, Cuba [Wooden Constructions in the Sea, Los Buchillones, Cuba]. El CaribeArqueológico, 7, 24–32.

Pendergast, D.M., Graham, E., Calvera, R.J., & Jardines, M.J. (2001). Houses in the sea: Excavation andpreservation at Los Buchillones, Cuba. In B.A. Purdy (Ed.), Enduring records: The environmental andcultural heritage of wetlands (pp. 71–82). Oxford: Oxbow Books.

Pendergast, D.M., Graham, E., Calvera R.J., & Jardines, M.J. (2002). The houses in which they dwelt:Excavation and dating of Taino wooden structures at Los Buchillones, Cuba. Journal of WetlandArchaeology, 1(2), 61–75.

Rippon, S. (2000). Exploitation and modification: Changing patterns in the use of coastal resources inSouthern Britain during the Roman and Medieval Periods. In A. Alberg & C. Lewis (Eds.), The risingtide: Archaeology and coastal landscapes (pp. 65–74). Oxford: Oxbow Books.

Rouse, I. (1992). The Tainos: Rise and decline of the people who greeted Columbus. New Haven, CT: YaleUniversity Press.

Sandweiss, D.H., Richardson, J.B., III, Reitz, E.J., & Maasch, K.A. (1996). Geoarchaeological evidencefrom Peru for a 5000 years B.P. onset of El Nino. Science, 273, 1531–1533.

Scholl, D.W., Craighead, F.C., & Stuiver, M. (1969). Florida submergence curve revised: Its relation tocoastal sedimentation rates. Science, 163, 562–564.

Schmitt, R.J., & Osenberg, C.W. (Eds.). (1996). Detecting ecological impacts: Concepts and applicationsin coastal habitats. San Diego, CA: Academic Press.

Sears, W.H., & Sullivan, S.O. (1987). Bahamas prehistory. American Antiquity, 43, 3–25.Seidemann, R.M. (2001). The Bahamian problem in Florida archaeology: Oceanographic perspectives on

the issues of pre-Columbian contact. Florida Anthropologist, 54, 4–23.Strobel, M. (2000). Die Schussenrieder Siedlung Taubried I: (Bad Buchau, Kr. Biberach): ein Beitrag zu

den Siedlungsstrukturen und zur Chronologie des frühen und mittleren Jungneolithikums inOberschwaben [The settlement Taubried I in the Schussenried culture in Bad Buchau, Kreis Biberach:

DOI: 10.1002/GEA GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5

STRATIGRAPHIC INVESTIGATIONS AT LOS BUCHILLONES

427

A contribution to the structures of settlements and to the chronology of the early and middle Neolithicage in Upper Swabia]. Stuttgart: K. Theiss.

Suelos de la Provincia Ciego de Avila. (1983). La Habana: Ministerio de Agricultura. Törnqvist, T.E., González, J.L., Newsom, L.A., Van der borg, K., De Jong, A.M.F., & Kurnik, C. (2004).

Deciphering Holocene sea-level history on the U.S. Gulf Coast: A high-resolution record from theMississippi Delta. Geological Society of America Bulletin, 116, 1026–1039.

Tucker Abbott, R. (1974). American seashells: The marine mollusca of the Atlantic and Pacific coasts ofNorth America. New York: Van Nostrand Reinhold.

Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail, P.R., Sarg, J.F., Loutit, T.S., et al. (1988). Anoverview of the fundamentals of sequence stratigraphy and key definitions. In C. Wilgus (Ed.), Sea-level changes—An integrated approach (SEPM Special Publication No. 42, pp. 39–45). Tulsa, OK:Society of Economic Paleontologists and Mineralogists.

Voorhies, B. (2000). Reconstructing mobility patterns of late hunter-gatherers in coastal Chiapas, Mexico:The view from the shellmounds. Anais do IX Congresso da Sociedade de Arqueologia Brasiliera.Retrieved June 16, 2004, from http://www.anth.ucsb.edu/faculty/voorhies/voorhies_2000.html

Wagner, E., & Tarble de Ruiz, K. (1975). Lagunillas: A new archaeological phase for the Lake MaracaiboBasin. Journal of Field Archaeology, 2, 105–118.

Walker, K.J., Stapor, F.W., & Marquardt, W.H. (1995). Archaeological evidence for a 1750–1450 B.P. higher-than-present sea level along Florida’s Gulf Coast. In C.W. Finkl (Ed.), Holocene cycles: Climate, sealevels, and sedimentation [Special issue]. Journal of Coastal Research, 17, 205–218.

Walker, K. J., Stapor, F.W., & Marquardt, W.H. (1994). Episodic sea levels and human occupation atSouthwest Florida’s Wightman Site. Florida Anthropologist, 47, 161–179.

Warmke, G.L., & Tucker Abbott, R. (1961). Caribbean seashells. Narberth, PA: Livingston PublishingCompany.

Warrick, R.A., Barrow, E.M., & Wigley, T.M.L. (Eds.). (1993). Climate and sea level change: Observations,projections, and implications. Cambridge: Cambridge University Press.

Waters, M.R. (1992). Principals of geoarchaeology: A North American perspective. Tuscon, AZ: Universityof Arizona Press.

Watters, D.R. (1998). Maritime adaptive strategies in the Caribbean archipelago. Revista de ArqueologiaAmericana, 15, 7–31.

Watters, D.R., Donahue, J., & Stuckenrath, R. (1992). Paleoshorelines and prehistory of Barbuda, WestIndies. In L.L. Johnson & M. Stright (Eds.), Paleoshorelines and prehistory: An investigation of method(pp. 15–52). Boca Raton, FL: CRC Press.

Weaver, C.E. (1989). Clays, muds, and shales. Developments in Sedimentology 44. Amsterdam: Elsevier. Wright, H.E., Jr. (1967). A square-rod piston sampler for lake sediments. Journal of Sedimentary Petrology,

37, 975–976.

Received November 5, 2004Accepted for publication June 23, 2005

PEROS, GRAHAM, AND DAVIS

GEOARCHAEOLOGY: AN INTERNATIONAL JOURNAL, VOL. 21, NO. 5 DOI: 10.1002/GEA428