ENVIRONMENTAL SETTING l ARCHAEOLOGICAL AND HISTORICAL SETTINGS

Written by: John Phillips and J. COZ Cozzi

ENVIRONMENTAL SETTING

The project area, shown in Figure 1, includes all of Escambia and Santa Rosa counties, in the extreme western panhandle of Florida. This region is bounded to the south by the Gulf of Mexico, and to the east by Okaloosa County, Florida. To the north and west the project area is bounded by the state of Alabama.

Physiography and Geology

Escambia and Santa Rosa counties are located within the East Gulf Coastal Plain physiographic province (Hunt 1974, Fenneman 1938). Sediments of the Citronelle Formation which blanket this province, were deposited during the Plio-Pleistocene epochs approximately one million years ago (Puri and Vernon 1964). During this epoch, high energy streams deposited upland sediments of sand, clays and gravels in alluvial fans which have coalesced on the coastal plain. These soft sandy deposits have been deeply eroded by streams which produced a relatively flat, yet highly dissected land surface.

The study area falls within two the physiographic subdivisions of the Coastal Plain: Coastal Lowlands and Western Highlands (Marsh 1966). The Coastal Lowlands are characterized by relatively non-dissected, nearly level plains that lie less than 100 ft above sea level. The Western Highlands region is a highly dissected, southward trending plateau. These Highlands are marked by rolling hills and extreme topographic relief that extends more than 280 feet above sea level.

During the Pleistocene Epoch (the last million years) a series of cold (glacial) and warm (interglacial) climatic episodes occurred. The glacial periods witnessed a substantial lowering (100-200m: 300-600 ft) of the sea level which increased stream velocity and caused the drainages to cut deeply into their valleys. The following rapid rises in sea levels associated with the interglacial periods flooded these deeply incised stream valleys and bays and greatly reduced stream velocity. This flooding filled the valleys with sediments of the soft Citronelle formation which continually erode from the uplands. In addition, during the interglacial periods drifting sediments in the Gulf of Mexico regularly formed barrier islands off the mainland which narrowed the bay mouths. Each succeeding interglacial period apparently produced a relatively lower sea level causing former peninsulas and barrier islands to be incorporated into the mainland. The sounds behind the former peninsulas and islands filled with sediments forming swamps in these long troughs. Approximate sea level changes during the past 40,000 years are shown in Figure 2.

Drainage Characteristics

The project area is drained by five river systems: Perdido, Escambia, Blackwater, Yellow, and East Bay. The Perdido is a relatively short river that drains the northwestern portion of Escambia County before emptying into Perdido Bay in the southwestern area of the county. The Perdido River and its tributaries are relatively narrow drainages with correspondingly narrow flood plains. The drainage system has a well developed dendritic drainage pattern. The stream valleys are V-shaped and have long and relatively steep slopes, and the stream terraces of the river and the larger tributaries are long and narrow. The slopes are long and steep and the ridge tops are narrow. In contrast, the slopes along the smaller tributaries are somewhat shorter, and the ridge tops are broad and nearly level. The ridges are composed of eroded sand deposits overlying poorly drained sandy clays or sandy clay loams. These sandy clay subsoils have produced numerous seepy areas and springs on the slopes and terraces, and very poorly drained sediments on the flood plains below. The Perdido valley is characterized by eroded sand hills with a natural forest cover of longleaf pine with an admixture of xeric hardwood forest, and mesic slope and bottomland hardwood forest.

The Escambia River is a large stream that drains the eastern side of the county. This river is essentially unaltered by marine land forms and is the largest and longest river in the project area. Within the work reach, the Escambia River is relatively broad with wide flood plains marked by extensive wetlands. The flood plains of the Escambia watershed are broader than those of the Perdido. The Escambia River and its tributaries are characterized by well developed dendritic drainage patterns. The stream valleys are V-shaped with long and relatively steep slopes. The stream terraces of the river and the larger tributaries are long and narrow, whereas the slopes are long and steep, and the ridge tops are narrow. The slopes along the smaller tributaries are somewhat short and the ridge tops are broad and nearly level. The ridges are composed of deep, loose and well drained sandy material overlying poorly drained sandy clays or sandy clay loams. Numerous seepy areas and springs are found on the slopes and terraces, and very poorly drained deposits occur on the flood plains. The Escambia River valley uplands are characterized by eroded sand hills with a natural forest cover of longleaf pine and hardwood forest. The Escambia River flood plains are characterized by cypress-gum swamp or bottomland hardwood forest.

Like the Perdido and Escambia drainages, the Blackwater stream valleys are V-shaped with long and relatively steep slopes. The stream terraces are long and narrow, the slopes are long and steep, and the ridge tops are narrow. The slopes along the smaller tributaries are somewhat shorter and the ridge tops are broad and nearly level. The ridges are composed of very deep, loose and excessively drained sandy material overlying poorly drained sandy clays or sandy clay loams. Numerous seeps and springs are found on the slopes and terraces. Very poorly drained sediments occur in the stream bottoms. The Blackwater River valley uplands are characterized by excessively drained sand hills covered in longleaf pine forests with an admixture of hardwoods on xeric hammocks. The narrow flood plains are characterized by titi swamps or bottomland hardwood forest.

The Yellow River flows from the northeast to the southwest into eastern Santa Rosa County before entering Blackwater Bay. Topographic relief both north and east of the Yellow River is more extreme than south of the river; the relief is also more extreme than in the Blackwater River valley. In Santa Rosa County the tributaries of the Yellow River are typically short, with steeply inclined drainage ways. Like the Blackwater drainage system, the Yellow River valley is characterized by excessively drained sand hills with numerous seeps and springs on the slopes and terraces. The natural vegetative community of Yellow River watershed is a sub-climax longleaf pine forest. The Yellow River valley also exhibits xeric hardwood hammocks, and mesic slope and bottomland hardwood forests.

The East Bay River drains an area south of the Yellow River valley that includes both excessively drained hills along the Yellow River drainage divide, and the extensive East Bay Swamp. This trough-like swamp is a former sound that is marked by vast flatwoods, sluggish streams and limited topographic relief.

The project area contains portions of Santa Rosa Sound, five large bays, and many of miles of bay and sound coast line. From west to east these bays are: Perdido, Pensacola, Escambia, East, and Blackwater. Topography near these lowland shorelines is characterized by low relief, sluggish streams and tidal creeks and some extensive marshes. Hardwood hammocks are present in the bay areas on better drained soils in areas protected from fire. There are two large peninsulas which jut into the bays: Garcon Point and Gulf Breeze peninsulas. The Garcon Point peninsula is characterized by extensive wetlands in the central and southern area and well drained sandy soils in the northern portion. These areas are separated by the well defined slope of the Penholoway Terrace which drops over 20m (60 ft) in less than a mile. Both sides of this triangular-shaped peninsula are eroding. The west shore erodes from winter storms and the east shore erodes from summer storms. The Gulf Breeze peninsula is a long and narrow land form stretching 30 miles west from the Holley/Navarre area to the City of Gulf Breeze on the western tip and ranges from 0.5 to 2 miles wide. This narrow peninsula is characterized by low relief, stabilized sand dunes and short spring-fed drainages. The dominant vegetation communities are pine flatwoods or sand pine scrub, however, hardwood hammocks are frequent along the shore. As on the Garcon Point peninsula, both shorelines are eroding from summer and winter storms.

Soils

This discussion of soils found within the project area is drawn from Weeks et al. (1980) and Carlisle (1960). Four general soil areas occur within the area. These soils are: 1) sandy, droughty soils; 2) well drained soils with a loamy subsoil; 3) well drained sandy and loamy soils; and 4) soils subject to flooding.

The sandy, droughty soils belong to the Lakeland-Troup association. These well drained to excessively drained soils are found in areas marked by broad, rolling sand hills dissected by long, narrow bottom lands; these bottoms are bordered by steep side slopes. These soils are poorly to moderately suited to cultivation and pasture; however, they are moderately well suited to timber production.

Areas dominated by well drained soils with a loamy subsoil are found in the Western Highlands and belong to the Red Bay-Lucy, Dothan-Orangeburg and Troup-Dothan-Bonifay associations. These soils occur on broad, nearly level to undulating hills and are well suited to most agricultural uses.

Areas dominated by poorly to very poorly drained sandy and loamy soils occur along the lower reaches of the major drainage ways. These sediments are found on level to gently sloping topographic features and are grouped within the Pactolus-Rutledge-Mulat association. These soils are poorly suited to cultivation, however, they are well suited to pasture land and for timber production.

The soils that are subject to flooding belong to the Bibb-Kinston-Johns and Dorovan-Pamlico associations. They occur on the flood plains and low streams throughout the major drainage systems of the Western Highlands. These soils do not respond well to most agricultural endeavors.

Climate

The northwest Florida area has a warm-temperate and humid climate (Weeks et al. 1980). While the summers are long and warm, the winters are short and very mild. The summer temperature averages about 80 degrees Fahrenheit; the average winter temperature is 54 degrees Fahrenheit. There are approximately 300 frost-free days in a typical year. Annual precipitation averages 65 inches and about half of this rain falls during the summer months. Prevailing winds blow from the south and southwest during the spring and summer; these winds generally blow from the north and northwest during the fall and winter.

Flora

The northwest Florida Gulf Coast is marked by vast longleaf pine forests (Braun 1950). The forests exhibit a number of vegetation communities. The mixed pine-oak forests of the uplands consist of an overstory of longleaf, sand and slash pine, post oak, black jack oak, turkey oak and live oak and an understory of saw palmetto, pineland threeawn, lopsided Indian grass and chalkey bluestem (Weeks et al. 1980). The stream terraces are covered in longleaf pine, live oak, red oak, hickory, magnolia and dogwood. The bottom lands adjacent to streams are forested in slash and long leaf pine, titi, dogwood, silver leaf bay and water tolerant oaks; the understory consists of gallberry, waxmyrtle, huckleberry and greenbrier. The larger bottomland swamps consist of an overstory of cypress, sycamore, sweetgum, blackgum, juniper, longleaf pine and water tolerant oaks and an understory of titi, waxmyrtle, ferns, greenbrier and muscadine.

Fauna

This environment supports a wide variety of wildlife. Among these are small and medium sized mammals such as the eastern cotton tail, common skunk, opossum, gray squirrel, gray fox, raccoon, armadillo and deer. Avian species characteristic of the area include several species of waterfowl, doves, woodpeckers, owls, hawks and other raptors, and a tremendous variety of small songbirds. Among the reptiles found in this environment are the box turtle, spotted turtle, diamond back terrapin, water snake, ring-necked snake, black snake, cottonmouth snake, pygmy rattlesnake, common anole, fence lizard and blue-tailed skink. Aquatic animals include sunfish, catfish, and several species of minnows.

Summary

Northwest Florida has a diverse environment that has undergone significant changes in the last 16,000 years, little of which is precisely understood. Modern conditions began about 3,000 years ago. The region is marked by considerable topographic relief, and a dendritic network of drainages that provided transportation corridors and ample water resources for small rural industries. The interior reaches are marked by marine-based, highly acidic sandy deposits that are generally not well suited for agriculture, although there are isolated expanses of upland sandy loam sediments that are agriculturally productive. The region is characterized by a natural fire sub-climax pine forest vegetative community that is interspersed with xeric pine-oak communities. These pine and pine-oak communities provided extensive timber resources with tremendous economic value.

ARCHAEOLOGICAL AND HISTORICAL SETTINGS

Previous Terrestrial Research

Formal archaeological investigations in northwest Florida began in the 1880s with a survey of shell midden sites along Florida's Gulf Coast (Walker 1885). Walker identified and excavated portions of shell middens and burial mounds in Escambia and Santa Rosa counties. Among these are Escribano Point (8SR2) and East Pensacola Heights (8ESl). The Escribano Point site, located at the head of East Bay, contained extensive midden deposits and produced several human burials. The East Pensacola Heights Site, located on Emanuel Point, contained burials within two sand mounds, and a shell midden. At the turn of the Twentieth century, C. B. Moore visited the northern Gulf Coast and investigated numerous sites within the region: the Santa Rosa Sound Site (8SR1), the Maester Creek Mound (8SR870) and Graveyard Point (8SR3) in Santa Rosa County, and the Fort Walton Temple Mound (8OK6) in nearby Okaloosa County, Florida (Moore 1901).

The next substantive archaeological research within the northwest Florida region was conducted by Columbia University under sponsorship of the National Park Service (Willey 1949). This extensive investigation of a 500 mile stretch of the Gulf Coast of Florida included surveys as well as test excavations at scores of sites. Many sites were tested in Escambia and Santa Rosa counties. In his monumental Archeology of the Florida Gulf Coast, Willey (1949) developed a prehistoric chronological framework which has formed the basis of all prehistoric chronologies for the region. This synthesis defined eight cultural periods and produced the first ceramic typologies for the Gulf Coast. Although the typologies have been refined over the years, the basic structure remains.

Following Willey's ground breaking work, archaeological investigations were undertaken in northwest Florida by Sears (1954), Fairbanks (1959; 1964) and Lazarus (1958; 1961). During the 1960s, Florida State University performed several limited investigations at sites in Santa Rosa and Escambia Counties. These investigations located the site of the presidio Santa Rosa Punta de Siquenza (8ES22), the "Second Pensacola", on Santa Rosa Island (occupied between 1719 and 1752) and documented structural remains and refuse pits from this settlement (Smith 1965). Other significant archaeological research undertaken in the region during this period includes Phelps' (1966) work in the central Florida Panhandle, and the survey of the Naval Live Oaks Reservation in Santa Rosa County by Tesar (1973). In addition, Percy (1974), Percy and Brose (1974), Brose (1984) and Sears (1977) generated refined chronologies for this area, especially for the Woodland and Mississippian stages.

Since 1983, the University of West Florida (UWF) has played a very active role in the archaeology of northwest Florida. This activity began with a reconnaissance level survey of the Escambia River valley (Bense 1983). Information generated by the Escambia River survey was used to define prehistoric interior-coastal settlement patterns. A multi-year survey conducted by UWF in the Blackwater River watershed recorded over 175 archaeological (Little, Curren, and McKenzie 1988a; Phillips 1989a; Penton 1991). UWF has also conducted reconnaissance level surveys in the Perdido River drainage (Little, Curren, and McKenzie 1988b; Little, Curren, McKenzie, and Lloyd 1988). Phillips and McKenzie (1993) used data derived from these drainage system surveys and other investigations to develop, test and refine a model of archaeological site locations in the Yellow River Marsh Aquatic Preserve. Phillips and McKenzie (1992a) and Phillips and Duncan (1993) developed and tested a model of archaeological site locations in Santa Rosa County using a computerized Geographic Information System (GIS). This study, designed as a limited test of GIS as an archaeological research tool, correlated and quantified specific environmental zones, primarily soil characteristics, that are associated with archaeological sites. A similar approach was used by Phillips (1995a) to develop a model of prehistoric site locations for Tyndall Air Force Base in nearby Panama City, Florida. Phillips and McKenzie (1997) developed a cultural resource management GIS for Pensacola that has been utilized a number of downtown development projects, including the Aragon Court Redevelopment project, and a cultural resource management GIS for the Naval Live Oaks Reservation in Santa Rosa County (Phillips and McKenzie 1998; Phillips 1999). Most recently Phillips, McKenzie and Wilson (2000) developed a cultural resources GIS for Santa Rosa County which is designed to assist county planners with their efforts to preserve significant historical resources.

A number of prehistoric sites have been tested by UWF. These include three Deptford sites in Escambia County: Hawkshaw (8ES1287) (Bense 1985), Main and Reus (8ES1378), 8ES1506, and several sites identified during the Escambia River survey (Bense 1994a). Archaeological testing at the Hickory Ridge cemetery documented three Late Mississippian period high status burials, and produced significant information on Bear Point phase mortuary practices (Phillips 1989b; 1995b). Limited testing has also been performed at a prehistoric Weeden Island midden site (8SR393) on Escambia Bay (Lee and Joy 1989). Other notable archaeological investigations conducted by UWF include survey and testing at Bernath Place (8SR986) (Memory 1991; Phillips 1992a, Bense 1994b). Archaeological excavations undertaken by Bense (1994b) at Bernath documented Santa Rosa-Swift Creek features and 17 burials.

A number of cultural resource preservation planning studies have been performed in the last few years for state, regional and local government agencies. These studies include an analysis of historic surveys in the northwest Florida region (Bense and Adams 1991) as well as state funded historic preservation plans for several counties and municipalities: the City of Pensacola (Bense 1989); Fort Walton Beach (Phillips (1992b); and Okaloosa (Phillips 1992c), Santa Rosa (Phillips and Bense 1990a), Escambia (Phillips 1992d) and Walton counties (Phillips and Anderson 2000).

Archaeological investigations have also been undertaken at numerous historic sites in northwest Florida (cf. Carruth 1989; Lee and Joy 1989; Little, Curren, and McKenzie 1989). Phillips (1993a) conducted extensive excavations at Arcadia (8SR384), a water-powered mill complex near Milton. This work documented three Antebellum mill structures, described an industrial artifact assemblage, and modeled the systems that powered the complex. Phillips (1993b) conducted a reconnaissance survey of water and steam powered mill sites that identified and described approximately 50 mill or mill related sites in Escambia, Santa Rosa and Okaloosa counties. A second mill survey, focusing on water-powered industries was completed by Phillips (1996). Building on this work, Phillips (1998) developed rural settlement models for the colonial and American periods. Much historic archaeological research has been conducted in Pensacola (cf. Bense 1985; 1989; Joy 1989a; 1989b; Joy and Lloyd 1988; Fabbro 1992; Stringfield 1992; 1995). Recent excavations in the Colonial community and inside the fortifications of Pensacola have documented First Spanish Colonial structures and features (ca. 1752-1763), British Colonial structures, fortifications, and features (ca. 1763-1781), and produced significant subsistence data. Ten years of archaeological research in Pensacola has produced a corpus of information, and a detailed description of the historical archaeological assemblages in Pensacola (Bense 1999).

Several cultural resource management compliance archaeological projects have been conducted in the area in recent years. These Cultural Resource Management (CRM) projects include surveys along a portion of the shore of Choctawhatchee Bay (Huston and Thomas 1984) and Santa Rosa Sound (Phillips 1984), the University of West Florida main campus (Phillips and Bense 1990b, Harris and Phillips 1995), mouth of the Perdido River (Phillips 1991) and the Tarkiln Bayou area (Curren, Mikell, Smith and Newby 1997), and interior areas of Okaloosa and Santa Rosa counties (Phillips 1989c, 1990; Phillips and McKenzie 1992b). Large scale surveys and limited test excavations have been undertaken on Eglin Air Force Base (Thomas and Campbell 1993), at Sandestin (New World Research 1985), and along a proposed pipeline corridor extending for about 120 miles through interior areas of northwest Florida (Phillips 1994; Phillips and McKenzie 1994). Test excavations in southern Okaloosa County have also been conducted in Fort Walton Beach at Pirate's Bay (8OK183) by Thomas and Campbell (1984) and Penton (1990). The Eglin investigations (Thomas and Campbell 1993) identified over 880 cultural occurrences, produced a predictive model of archaeological site locations, and provided a comprehensive synthesis of the archaeological research undertaken in northwest Florida.

Previous Underwater Research

Prior to the 1970's work on underwater sites was conducted periodically by the Army Corps of Engineers and private salvors to remove obstructions to navigation and to salvage ship's equipment. During the most recent obstruction removal the US Navy removed the mast and did other work on the remains of the small armed schooner Preble lost during the Civil War, in Pensacola. In 1973 the National Park Service (NPS) conducted a preliminary remote sensing survey of waters under its management responsibility around Pensacola Pass, while at the same time contracting with Florida State University's Department of Anthropology who conducted a cultural resource survey of NPS lands. Shipwreck sites in littoral waters were located, assessed, and assigned Florida Master Site File numbers. These sites were predominantly nineteenth-century wreck sites or ballast scatters and ranged from the Norwegian bark Catharine sunk in Gulf waters in an 1894 gale to the tugboat Sport sunk in Santa Rosa Sound during the 1906 hurricane. Concurrent with this project G. Norman Simons, at that time director of the Pensacola Historical Museum, prepared a listing of known Pensacola shipwrecks from a variety of historical and archival sources. Louis Tesar, who headed the Florida State University (FSU) investigation, noted that he had witnessed and heard of many instances of unauthorized salvage activities. In 1979 the National Park Service conducted a second remote sensing survey off Perdido Key and determined that four of the targets detected in 1973 were still buried.

In 1986 the Army Corps of Engineers, Mobile District, surveyed the harbor channel and the turning basin off from Naval Air Station, Pensacola, for the Navy Strategic Homeport Program in which the Navy planned to base the aircraft carrier USS Forrestal at Pensacola. The survey detected 173 magnetic anomalies, of which 56 had corresponding sonar images. Of these a dozen were selected for ground truthing. The next year Tidewater Atlantic Research (TAR) supervised Navy divers from the Navy Experimental Diving Unit in Panama City, Florida, who relocated and assessed seven of the twelve Corps targets as well as two others identified by local divers. Only one historically significant site was identified. This was a site provided by a local informant and proved to be the remains of the Convoy, which was a steamboat sunk in the 1860s and which has undergone obstruction removal work. Artifacts from the shipwreck were collected prior to this work by a local diver who donated them to the Pensacola Historical Museum for curation and display.

In 1988 and 1989 several sites were examined by archaeologists and interest in the underwater archaeology of Pensacola reach a new high and continued to build momentum which has lasted until the present. The Florida Bureau of Archaeological Research (BAR) examined a site reported by Milton resident Warren Weeks which proved to be a nineteenth-century lumber schooner in a remarkable state of preservation. The Blackwater River Bethune Schooner site (8SR985) was initially documented by BAR personnel in 1988 and the next year a volunteer crew lead by David R. Baumer of East Carolina University's Maritime History and Underwater Archaeology Program recorded the site in detail and produced a report of investigations that recommended further examination.

In 1988 the University of West Florida Archaeology Institute conducted a survey of Deadman's Island in Gulf Breeze, Florida which revealed several sites associated with maritime activity including a marine railway for servicing ships and an abandoned colonial-period vessel. The ship remains were in extremely shallow water that allowed its recording without the use of specialized diving equipment. In 1989 a joint BAR/UWF field school investigated the vessel's remains and determined it to be either the HMS Florida or the HMS Stork.

In 1989 a contract archaeology firm conducted a survey in the Blackwater River at the junction of Pond Creek, but did not find any significant sites. That same year the University of West Florida's first class in underwater archaeology recorded the site of Jack's Wreck (8ES1900) on Perdido Key. Historical research indicates that the vessel may be the remains of the fishing vessel Lucky Strike abandoned in that area in 1958, although further research is required to verify this conclusion.

In the 1990s several chance finds and the institution of methodical survey projects produced several significant finds. In 1990 a bronze howitzer was dredged up by the Army Corps of Engineer's Navy Strategic Homeport Program. A survey of the area did not turn up any further evidence of significant remains, although ship timbers had been brought up in the dredging that produced the howitzer. Renovation of a pier at Naval Air Station, Pensacola, as part of the Navy Strategic Homeport Program that same year produced copper sheathing and ship timbers associated with a large caisson which had been intentionally sunk during wharf-building in the 1830's. The caisson was removed during dredging operations with efforts made to document the remains to produce a scale model.

The Florida Bureau of Archaeological Research (BAR) instituted a systematic survey of Pensacola Bay in 1991. The National Oceanographic and Atmospheric Administration (NOAA) Coastal Zone Management program granted funding for the Pensacola Shipwreck Survey. The survey was supported with office space by the Historic Pensacola Preservation Board (HPPB), while many Pensacola businesses and individuals provided information, services, and volunteer assistance. The survey staff relied on local informants as well as employing remote-sensing tools. The project recorded over three dozen vessels many of which had been abandoned rather than wrecked. A second field season was conducted in 1992 which relied more heavily on remote sensing equipment. Among the sites found was the Emanuel Point Ship (8ES1980), which has been identified as from the 1559 fleet of Tristán de Luna who attempted the first settlement of what is today Florida. This site is Florida's oldest shipwreck and was examined between 1993 and 1995 and then again 1997. Other finds include the Santa Rosa Island Wreck and Hamilton's Shipwreck which are currently the subjects of investigation by University of West Florida's Archaeology Institute. In 1992, the Pensacola Shipwreck Survey documented the remains of the turn-of-the-century battleship USS Massachusetts, sunk west of Pensacola Pass, to assist in its designation as Florida's fourth State Underwater Archaeological Preserve. Massachusetts is the nation's oldest existing battleship.

The University of West Florida has take a leading role in Pensacola's shipwreck archaeology since it came on board as a partner with BAR in the 1997 excavation of the Emanuel Point Ship. In 1998 UWF signed a Cooperative Research and Development Agreement (CRADA) with the US Navy to conduct a remote sensing survey of portions of Pensacola Bay. This survey produced over 300 magnetic anomalies and more than 300 side-scan sonar targets. These targets have formed the basis for the current fieldwork to ground truth targets to determine which are historically significant. Also in 1998 UWF documented the remains of the Norwegian ship, Catharine, following reports that local divers had been finding and in cases removing artifacts. This work was undertaken at the request of the National Park Service. In 1999 UWF excavated the bow of the Santa Rosa Island Wreck and has determined that the best historical candidate for this vessel is the Nuestra Senora del Rosario y Santiago Apostol, which was lost in a hurricane in 1705. The current UWF investigations are a part of the systematic survey of Pensacola Bay begun by BAR in the early 1990's.

Prehistoric Chronology

In combination, the archaeological investigations undertaken over the last 100 years in northwest Florida have produced much data on the prehistory and history of the region, and a picture of human adaptation to the northwest Florida area has emerged. Table 1 delineates the cultural chronology of the region and a brief overview of the culture history of the area is presented below. For a more complete delineation of northwest Florida cultural chronology, the reader is referred to Bense (1989, 1994a), Milanich and Fairbanks (1980), Milanich (1994), and Thomas and Campbell (1993).

Paleoindian Stage (ca. 12,000 - 8,000 B.C.)

The Paleoindian Stage, representing the earliest occupation of the Western Hemisphere, began during the late Pleistocene (Great Ice Age) and ended about 8,500 B.C. (Bense 1994a:38). These first human occupants of North America probably migrated from northeast Asia during the latter stages of this last glacial period. After crossing the Bering Land Bridge, a wide savannah that once connected Siberia with Alaska, these migrants apparently moved southward through the Canadian ice sheet along an ice-free corridor into the upper Great Plains. From the plains they probably followed eastward trending rivers into the central Mississippi valley before spreading east and south along the Ohio, Tennessee and Cumberland river valleys (Bense 1994a:38).

The Paleoindian stage is divided into three periods, Early, Middle and Late, primarily on the basis of changes in projectile point styles. The hallmark of the early and middle Paleoindian periods is the lanceolate shaped projectile point with a convex or straight haft element and base (Bense 1994a: 41-42). These artifacts are long, thin bifacially worked blades, made of fine grained chert, that often exhibit a flute, or long shallow flake scar, on each face (see Figure 3). Early and middle Paleoindian sites in north Florida are marked by the Suwannee and Simpson projectile point types, and an expedient lithic tool kit of unifacial scrapers, adzes, blades, spokeshaves, flake knives, and ground stone implements (Milanich 1994:48-51). Late in the Paleoindian stage, climatic events evidently triggered changes in settlement and subsistence. These developments are reflected in late Paleoindian settlement systems and assemblages. The Late Paleoindian assemblage includes small serrated projectile points, and the addition of woodworking implements (Bense 1994a:55).

Early and Middle Periods (ca. 13,000 - 8,500 B.C.)

Paleoindians probably came to north Florida near the end of the Pleistocene, approximately 12,000 years ago (Milanich 1994: 37). Climatic conditions in Florida were substantially different from today. The region was drier and several degrees cooler, sea levels were about 160 ft lower, and the Gulf Coast shoreline was located well south of its present position. Modern coastal areas, which are now flat, low and wet, were formerly dry uplands marked by extreme topographic relief. Inland drainages, springs, streams, and wetlands were virtually non-existent, and the water table was much lower. Potable water during the late Pleistocene was scarce, and it was available in the rain-fed water holes and lakes, and deep sinkholes fed by springs (Milanich 1994:39). These water holes, lakes and deep sinks were most plentiful in the karstic limestone deposits of north peninsular Florida, from near Tampa north to the Chipola River; these limestone deposits also extend off the Gulf Coast. Most of these sinks and lakes are now inundated.

Recent research by Dunbar, Waller, Webb and their associates (cf. Waller and Dunbar 1977; Dunbar 1981; Dunbar and Waller 1983; Dunbar et al. 1989) has contributed significant settlement and subsistence information about the early and middle Paleoindian periods in Florida. These researchers have demonstrated that most Paleoindian sites are located in these areas of karst topography: the Santa Fe-Ichetucknee River basin; the Aucilla-Wascissa River system; the Steinhatchee River; the Oklawaha River-Silver River basin; the Withlacoochee River; the Hillsborough River drainage; the Chipola River system; and the St. Johns River (Milanich 1994: 43). Many of these sites are now under inundated. The Page/Ladson site on the Aucilla River has yielded radiocarbon dates of 10,000-7500 B.C. for strata producing Paleoindian materials (Milanich 1994: 46). At Harney Flats near Tampa, Daniel and Wisenbaker (1987) excavated a Paleoindian site situated on a ridge divide between two river valleys that overlooked a savannah. This almost completely undisturbed Paleoindian settlement yielded thousands of artifacts that provided data on the production sequence for Suwannee points, and enabled these researchers to delineate special activity areas. At the Silver Springs Run Site, archaeologists recovered mammoth bones in association with a stemless point and chert flakes (Hoffman 1983:83-87).

Dunbar and Webb's work indicates that water was the crucial element in Paleoindian settlement. This "oasis model" suggests that water holes, sinks, and lakes provided the Paleoindians
with relatively rich natural communities that supported a diversity of plant and animal resources (Milanich 1994:41). The current theory of Paleoindian subsistence is that it was readily adaptable to unfamiliar and rapidly changing environments. The primary subsistence activities were hunting, fishing, and collecting a wide range of faunal and floral resources (Bense 1994a:44).

Little is known about the Paleoindian occupation in northwest Florida, and only one middle Paleoindian site has been recorded in extreme northwest Florida. The multi-year Eglin research failed to produce any early Paleoindian sites, although these researchers note that several Paleoindian sites have been reported in a submerged context near the base (Thomas and Campbell 1993:503). Paleoindian markers have been discovered by collectors in northwest Florida, from the Chipola River, St. Andrew Bay, Choctawhatchee Bay and Escambia Bay. Most sites of the early and middle Paleoindian periods probably are submerged or may be deeply buried in the bays and interior flood plains.

Late Paleoindian Period (ca. 8,500 - 8,000 B.C.)

The late Paleoindian period perhaps represents a transitional era between the Paleoindian and Archaic stages. Marked by the Dalton projectile point style, this period exhibits changes in technology is interpreted by most archaeologists as reflecting a change from the cooler, wetter environment of the late Pleistocene to the warming and drying climate of the Holocene (Milanich 1994). Dalton-like points are shown in Figure 3. Characterized by a dramatic increase in the number of sites in comparison to the preceding periods, the late Paleoindian period is thought to be a time of increasing populations with the establishment of many new settlements. These settlements are found not only in the same locations as preceding periods, but in upland areas as well. Dalton peoples apparently retained a hunting and gathering economy, although, the plant and animal resources were modern. Recent investigations in Escambia, Okaloosa, and Santa Rosa counties have produced Dalton sites in upland settings near lower order drainages (cf. Phillips 1989a; Phillips and Bense 1990a; Penton 1991a; Phillips and McKenzie 1992a; Phillips 1992c, Phillips 1994).

Archaic Stage (ca. 8,000-1000 B.C.)

The Archaic stage occurred between 8000 and 1000 B.C. (Bense 1994a: 62). The longest stage of cultural development in the region, this stage is marked by notched and broad-bladed projectile points, containers, and ground and polished stone implements. Among the important cultural developments that began during this stage are the initial construction of mounds and earthworks, the establishment of long distance trade, and the development of large settlements and sites (Bense 1994a). The Archaic stage began at the outset of the Holocene climatic epoch, and ended with the beginning of modern climatic conditions. Global warming peaked midway through the Archaic, and has been cooling ever since. Concomitant with the changes in climatic conditions was a steady rise in sea level, and a change from deciduous hardwood forest communities to a fire sub-climax pine forest community.

The Archaic stage witnessed an increase in the number and diversity of artifacts produced in comparison to the preceding Paleoindian stage (Bense 1994a: 64-65). Projectile points, the most sensitive diagnostic artifact type, changed from the lanceolate-shaped type of the Paleoindian stage to triangular shaped artifacts with notched or stemmed bases (see Figure 3). In addition, ground and polished axes, celts, and ornamental items were added to the Archaic lithic inventory. Trade networks were established in which personal ornaments and raw material were exchanged; these ornaments also became high status markers. Containers of wood, stone, and later pottery, first appeared during this stage. Other Archaic artifacts include textiles and baskets, wooden dugouts, tools, and adornments.

Research indicates that by the Archaic stage, people had developed a diversified economy, and a wide variety of animal and plant resources were exploited (Bense 1994a:65). There is some evidence that early forms of agriculture began during the Archaic stage. Archaic settlement systems utilized seasonal base camps and short term special use camps.
The Archaic stage in the East Gulf Coastal Plain is divided into three periods (Bense 1994a). The Early Archaic period existed from 8000-6000 B. C.. The Middle Archaic period lasted from 6000-4000 B.C. The Late Archaic began around 4000 B.C. and ended in 1000 B.C.

Early Archaic Period (ca. 8,000 - 6,000 B.C.)

At the beginning of the Early Archaic period, conditions were substantially different from today. Sea level was about 90 ft lower, the climate was becoming warmer and drier, and the region was covered in hardwoods (Bense 1994a: 65). The archaeological record indicates a continued reliance on a traditional hunting and gathering economy; however, there is a hint of increased specialization. The Early Archaic assemblages are more diverse with increasingly specialized stone tools performing a variety of tasks (Milanich 1994). The assemblage includes chipped stone knives, scrapers, and adzes, and a variety of cutting and scraping implements produced on blades struck from prepared cores. A specialized bipolar industry and ground stone celts also were developed during this period. The Early Archaic is marked by small side and corner notched projectile points, and by the invention of the spear thrower or atlatl. These technological developments are thought to have enabled hunters to more efficiently kill the small fast moving game that inhabited the post-Pleistocene environment. The Early Archaic settlement pattern evidently involved seasonal movements of small family-based groups who occupied small short-term camps for part of the year. During the fall, when food was more plentiful, these small groups may have gathered at larger base camps (Bense 1994a: 72).
Middle Archaic Period (6,000 - 4,000 B.C.)

This period marks the beginning of a shift from the dispersed settlement pattern of the preceding period to a system of more nucleated flood plain base camps with numerous, smaller satellite camps (Bense 1989). These changes in settlement maximized the resources of the forest and coincided with the climatic episode known as the Altithermal; a period in which post-glacial temperatures peaked, and rainfall diminished. These economic adaptations involved hunting, fishing and gathering technologies that exploited a few primary resources within the interior forest (Caldwell 1958; Dye 1980). Artifacts associated with this period include broad-bladed, stemmed projectile points, specialized tools such as microliths, burins, large chopping implements, and an array of expedient tools.

Middle Archaic settlement and subsistence patterns evidently were similar to those of the Early Archaic (Bense 1994a:82). Macro bands of extended families probably occupied part or all of a watershed. These groups apparently separated into smaller family groups who moved from campsite to campsite through the territory. For a part of the year these groups may have coalesced at larger base camps. Toward the end of the Middle Archaic, in the eastern part of the Florida panhandle, there may have been a change from this dispersed settlement pattern to one of more sedentary existence (Bense 1994a:82). Exceptionally large sites developed with thick midden deposits and numerous features.

Recently, a significant early middle period cemetery in east central Florida was excavated (Stephenson 1985). The Windover Site (8BR246) has yielded extensive information about the biology, diet, textile industry and environment of this period. Religious paraphernalia associated with Windover burials suggests a well-developed religious ceremonialism. Windover promises to add much new information about Archaic lifestyles.

Middle Archaic sites in the northwest Florida area are relatively rare, and few have been recorded in the area. Eglin also produced very few Middle Archaic sites (Thomas and Campbell 1993:514-515). Like the preceding period, the Middle Archaic is probably represented in northwest Florida by isolated finds or low density occurrences.

Late Archaic Period (ca. 4,000 - 1000 B.C.)

The Late Archaic was a time of innovation that eventually brought about a new stage of cultural development. These innovations include the use of containers of stone and clay, plant food cultivation, increased trade, and mound building (Bense 1994a:85). The environment was changing, from the hot dry climate of the Altithermal, to cooler and wetter conditions; sea level also began to stabilize. Evidently marine resources became more abundant. Subsistence strategies and technologies reflect the beginnings of an adaptation to these resources; it is during this period that coastal shell middens like the Meig's Pasture Site in Okaloosa County began to appear in this area (Curren 1987). Evidently there were increases in population and in sedentism. Base camps, strategically located near the developing wetlands to target marine resources, begin to appear with greater frequency in the Late Archaic.

The projectile point styles that mark the beginning of the Late Archaic include broad-bladed, stemmed projectile styles that are very similar to those of the preceding Middle Archaic period. Ground and polished stone tools and ornaments continued to be developed in this period. Tiny ornaments, often carved in to the shape of animals, were produced. Another innovation in stone utensil use was the development of steatite cooking vessels. Ceramic utensils are added to the assemblage toward the end of the Late Archaic.

The pre-ceramic Late Archaic remains problematic. Continued use of broad-bladed projectile points makes recognition of these components difficult to distinguish from the Middle Archaic. Nonetheless, the pre-ceramic Late Archaic occupation of northwest Florida appears to be somewhat limited. Again, like the preceding period, the pre-ceramic Late Archaic is probably represented in northwest Florida by isolated finds or low density occurrences (Thomas and Campbell 1993:517). This area was an excessively drained upland ridge with limited water and a very sparse environment. Scores of these low density pre-ceramic sites have been reported in the Escambia and nearby Blackwater river valleys (Bense 1983; Little et al. 1988; Phillips 1989a).

he end of the late Archaic, about 1,000 B.C., signals the introduction of fiber-tempered ceramics of the Norwood series (Milanich 1994:95). This fiber tempered ware was later replaced by early sand tempered ceramic types (see Figure 4). These occurrences indicate significant changes among late Archaic cultures in the greater southeast. Among these are further increases in population, an increase in the number of dense village middens, and even more reliance on marine resources. In addition, there is some evidence of regional interaction with other cultures such as the Poverty Point complex of the lower Mississippi valley. The northwest Florida variant, termed Elliott's Point, includes fired clay balls, a microlith industry and other Poverty Point-like expressions (Thomas and Campbell 1993:522). The Elliott's Point complex initially appeared in northwest Florida around 2,000 B.C. and reached its classic form by 1,000 B.C. (Thomas and Campbell 1993:522-541). The complex is marked by a distinctive assemblage of baked clay balls, microliths, and exotic items that suggest connections with the lower Mississippi Valley; the Norwood ceramic series was added to the assemblage at approximately 1,000 B.C. Thomas and Campbell (1993:528-529) believe that Elliott's Point settlement mirrors Poverty Point. They see a general pattern of small sites that are aligned with around regional centers. Regional centers have mounds or ring middens organized into horseshoe, semicircular or oval shapes.

Woodland Stage (1000 B.C.-A.D. 1200)

The Woodland Stage is the third major stage of cultural development in eastern North America. The climate and major forest zones during the Woodland Stage were much as today. But the sea-level continued to fluctuate until approximately 400 B.C. (Bense 1994a: 110). The Woodland stage is marked by population increases in the interior river valleys, and along the coast an increased use of pottery, and more elaborate ceremonials and mortuary rituals (Bense 1994a). This stage also saw more plants cultivated in the northern southeast, more burial mounds constructed, and increased trade networks. While none of these traits developed in the Woodland stage, they became more common, and increasingly complex during the stage.
The Woodland Stage is divided into three periods. The Early Woodland period in the East Gulf Coast region probably began about 1000-500 B. C. and ended around A.D. 200 (Milanich 1994 111-114). The Middle Woodland period lasted from A. D. 200 to approximately A.D. 500 (Thomas and Campbell 1993: 491), whereas the Late Woodland period began around A.D. 500 and lasted until perhaps A. D. 1200 (Bense 1989).

Early Woodland Period (ca. 1000 B.C. - A.D. 200)

In north Florida and southern Georgia, the Early Woodland period is represented by the Deptford culture (Milanich 1994:111-114). The main aspects of the culture include a coastal-riverine subsistence base, a religious complex and a base camp-satellite camp settlement pattern (Bense 1989). This subsistence system evidently correlates with the emergence of modern climatic conditions, and the development of swamps, wetlands, and marine resources. The base camps are marked by the presence of shell middens, and are located almost exclusively along the coast within maritime hammocks near brackish or fresh water (Bense 1985; Milanich 1994:116). The satellite camps are small special activity sites that are often located within interior stream valleys near major drainages. This settlement pattern began in the late Archaic, and marked a definite shift in settlement toward the coastal lowlands. Deptford base camps tend to be strategically located in ecotonal settings. Living near the environments of both forest and the marsh, each supporting a rich and diverse natural community, provided a stable, multifaceted resource base. No doubt this diversity was the strong draw that brought the Deptford people to the coast. The settlement and subsistence pattern established by the Early Woodland Deptford culture was continued relatively unchanged for the next 2,000 years along the eastern Gulf and Atlantic coasts.

Deptford culture is characterized by a series of paddle stamped, sand tempered ceramics and minority amounts of net and fabric impressed types (see Figure 4). Among the Deptford vessels are large, deep, and cylindrically shaped cooking pots with rounded bottoms, and short cylindrical vessels; some Deptford vessels also have distinctive podal supports. Stone tools on Deptford sites are rare (Milanich 1994:126). Large stemmed projectile points and medium sized triangular points, bifacially worked tools, small blades, and expedient flake tools have been recovered in very limited numbers. The lithic assemblage also includes ground and polished celts, limestone and sandstone grinding implements, hammer stones and whetstones. Bone tools, including points, awls, flakers, pins, and gouges, are occasionally recovered from Deptford sites. Oyster and clam shells were used as ladles, spoons, cups and dippers, whereas whelk shells were utilized as picks or axes.

Excavations at two Deptford sites in northwest Florida, Hawkshaw (8ES1287) and Pirate's Bay (8OK183), have produced much information about Deptford subsistence and settlement (Bense 1985, Thomas and Campbell 1984). Both are coastal sites that provide evidence that Deptford people exploited a wide range of local marine and terrestrial food resources. In addition, trade ceramics from the lower Mississippi Valley Tchefuncte culture, and Swift Creek cultures of Georgia were recovered from both sites, indicating that a wide trade network existed during this time. Mikell et al. (1989:216) describe a strong Deptford presence on Tyndall, that supports the model that Deptford culture was a coastal oriented society in northwest Florida (Milanich 1994:112).

Middle Woodland Period (ca. A.D. 200 - 500)

The Middle Woodland period in northwest Florida developed out of the Deptford culture (Bense 1994a:158). The period is recognized by ceramic changes and an increase in sites. Settlement and subsistence patterns remained fairly consistent with that of the Deptford culture. In the interior areas, flood plain base camps are found in the stream valleys and temporary camps are found in the uplands. An analogous pattern is found along the coast: Large shell middens, representing lowland base camps, sometimes made in the shape of horseshoes and rings, are found in the lowlands. Surrounding the base camps are several satellite sites. These small special activity site occur along the coast and in the interior (Bense 1994a:159). There are indications of settlement clusters that consist of one large base camp and several satellite sites. In the Santa Rosa-Swift Creek area, the large site is commonly a ring or horseshoe shaped midden with an associated burial mound (Bense 1994a:159).

This Santa Rosa-Swift Creek culture is marked by two ceramic series: Swift Creek and Santa Rosa (see Figure 4). The Swift Creek series is a complicated stamped pottery that apparently developed out of the Southern Appalachian tradition (Bense 1994a:158). Swift Creek ceramic designs consist of curvilinear elements such as scrolls, concentric circles, teardrops, and spirals. Santa Rosa pottery, evidently a variant of the Lower Mississippi Valley-northern Gulf Coast Marksville series, represents a continuation of the Gulf Tradition: incised, punctated and rocker stamped design elements. Santa Rosa-Swift Creek vessel shapes include open jars, and bowls with podal supports and notched and scalloped rims. The Santa Rosa-Swift Creek tool kit includes stone, bone, and shell implements. The lithic assemblage includes expanded stemmed projectile points, bifacial knives, spokeshaves, flake scrapers, limestone, sandstone and chert abraders, and pitted anvil stones. Bense and Watson (1979) recovered bone awls, flakers, polished pins, and bone scrapers at 8BY73; cut and polished carnivore teeth and mandibles are also known from village middens (Milanich 1994:145). Although shell tools are rare, whelk ladles and cups, picks, hammers and pounders also occur on Santa Rosa-Swift Creek sites.
The Santa Rosa ceramics are viewed by some researchers as having different uses in different areas (Phelps 1969). In the Santa Rosa culture area, which may have centered west of Pensacola, Santa Rosa ceramics are believed to have been used as every day ware, but east of the Apalachicola River these ceramics were used only as mortuary wares. In essence, the Middle Woodland culture in northwest Florida evidently was the product of a blend of influences: Marksville from the lower Mississippi Valley, Hopewell from the Ohio Valley and the complicated stamped tradition from the Georgia area.

The socio-religious aspect of this culture has been defined as the Green Point complex (Sears 1962). It has been suggested that the Middle Woodland mounds may have been used for more than mortuary activities (Penton 1974). Some mounds have been described as flat topped, others had ramps, whereas some evidently had structures on the summits. The major stimulus for the socio-religious complex was participation in the Hopewellian interaction sphere. Santa Rosa peoples, along with contemporaneous Woodland cultures to the west, north and east participated in an exchange of exotic items such as copper, mica, conch shells, ear spools, and ceramics and began to construct burial mounds (Bense 1989; Milanich 1994:133-141). The religious ceremonialism associated with the complex that developed during this period contained themes which continued throughout prehistory. Socio-political specialists began to emerge as specialists, and sacred materials became separated from the secular environment. Sacred paraphernalia was interred with this specialist class of high status individuals in burial mounds.
Recent investigations at the Bernath Place site (8SR986), located in nearby Santa Rosa County, have provided much information about Santa Rosa/Swift Creek subsistence, settlement, and socio-political and religious organization (Phillips 1992b; Bense 1992; 1994b). Bernath Place is a coastal site with remnants of an intact Santa Rosa-Swift Creek and early Weeden Island midden, and domestic features that provide evidence that a wide range of local marine and terrestrial food resources were exploited. Radiocarbon dates ranging between A.D. 350 and A.D. 590 were obtained from two sealed Santa Rosa-Swift Creek features (Phillips 1992b). Field work also documented the remains of a ring midden surrounding a semi-circular central plaza swept free of debris; this plaza contains numerous burials. Bense (1994b) suggests that Woodland ring midden sites, such as 8SR986, functioned as socio-political centers, each with a semi circular central plaza swept free of debris. She further suggests that the historically documented square grounds of traditional Muskogean societies derived from these semi-circular plazas.

Late Woodland Period (ca. A.D. 500 - 1200)
The late Woodland period in northwest Florida has been traditionally called Weeden Island (Willey 1949). Recent research has defined this construct so that "Weeden Island" now refers to several distinct regional cultures that showed the same basic ceremonial complex. This complex may have been associated with specific social and political patterns (Milanich 1994). Not all Weeden Island traits are found in all areas; these cultures are referred to as Weeden Island "related" or Weeden Island "period" cultures. Recent research by Bense (1994a) and Thomas and Campbell (1993) indicates that the late Woodland Period lasted until approximately A.D. 1200 in extreme northwest Florida. These long lasting late Woodland cultures appear to have occurred all along the Atlantic and Gulf coastal areas except at the mouths of major rivers such as the Mobile-Tensaw, Savannah and Apalachicola.

The Weeden Island culture was originally divided into two chronological periods, I and II based on village midden ceramic assemblages (Willey and Woodbury 1941, Willey 1949). Weeden Island ceramic types are shown in Figure 4. Weeden Island I was marked by Swift Creek Complicated Stamped ceramics and the Weeden Island Incised and Punctate types. Weeden Island II was identified by a lack of complicated stamping, a reduction in the frequencies of the Weeden Island Incised types and the appearance of check stamped ceramics. Mikell et al. (1989:219-229) refined the Weeden Island chronology using seriation to define three periods of Weeden Island on Tyndall Air Force Base in Bay County that may have some validity for the Pensacola area: Weeden Island I; Weeden Island II; and Terminal Weeden Island. Weeden Island I is characterized by late varieties of Swift Creek Complicated Stamped and St. Andrews Complicated Stamped ceramics, small quantities of Weeden Island incised and punctated types, and 50 to 80 percent plain sand tempered wares. Weeden Island II is marked by high percentages (upwards to 43 percent) of Wakulla Check Stamped ceramics, a diminished percentage of plain sand tempered ceramics (ca. 50 percent), and an increase in the frequencies of Weeden Island incised, punctated and other surface treatments; complicated stamped varieties occur in very low numbers. Terminal Weeden Island is dominated by Wakulla Check Stamped, decreased frequencies of plain sand tempered wares, and very limited numbers of Weeden Island incised and punctated types.

Recent investigations in northwest Florida have suggested that there was change in settlement during the Weeden Island period (Bense 1989). Village midden and mound sites are often found in different settings than earlier Middle Woodland period sites. At these new locations, the ring midden patterns continued, and rings, horseshoes or rectangles have been documented. New burial mounds, some with ramps and flat summits were constructed near the large middens. The mounds contained the standard assemblage of sacred paraphernalia found throughout the culture area. Subsistence studies suggest that these peoples continued to exploit shellfish, fish, deer, and nuts as primary food resources. There appears to be some continuity between Santa Rosa-Swift Creek and Weeden Island occupations (Mikell et al. 1989:218) . Not only are both cultural expressions found in the same coastal environmental settings, but these cultures exploited similar marine resources.

In the latter part of the late Woodland period, new settlements appear in the upper reaches of the bay systems (Bense 1989). The patterns of settlement consisted of several small communities organized around a nearby mound. The appearance of these new settlements is interpreted as a reflection of an increase in population and resultant pressures on the food resources in the lower bays where only larger settlements once existed.

Mississippian Stage (A.D. 1200 - 1700)

Research suggests that this culture evolved in the Apalachicola River valley ca. A.D. 1000 (Scarry 1980, 1981, 1982). This Apalachicola-Fort Walton culture, stimulated by cultural connections with emerging Mississippian groups upriver, evidently evolved in place out of the preceding Weeden Island culture. These contacts apparently brought new ideas to the indigenous population for organizing increasingly larger societies and more intensive and efficient agriculture. Mississippian peoples developed complex chiefdoms based on maize agriculture and a redistribution tribute system. Politico-religious centers such as the Lake Jackson site near Tallahassee developed (Jones 1982). This center had several mounds and was associated with a ruling class.

In coastal areas of northwest Florida , the Mississippian way of life was different from that of the agriculturally-based Apalachicola-Fort Walton (Bense 1989). Stowe (1984) describes this coastal Mississippian culture as the "Pensacola variant". The area included by the "Pensacola variant" encompasses the northern Gulf Coast from Choctawhatchee Bay west to the Pascagoula River, up the Tombigbee River to Jackson, Alabama and up the Alabama River to Camden, Alabama. Two phases are identified: Bottle Creek (A.D. 1200 - 1450) and Bear Point (A.D. 1450 - 1700). Each phase has distinctive ceramic assemblages, although there is an overlap of types (see Figure 4). The main source of Mississippian influence and cultural interaction apparently derived from the north (Moundville). In essence, the Pensacola culture is one of many coastal Mississippian societies distinguishable by ceramic assemblages and settlement strategies (Knight 1980; Tesar 1980).

The Woodland settlement pattern and subsistence system continued in the Mississippian stage in northwest Florida. Large settlements, located on coastal hammocks, were surrounded by many smaller satellite camps, often in a variety of coastal and upland environments. Coastal Mississippian subsistence continued the Woodland strategy and was based on hunting, gathering and fishing. Agriculture is not considered to have been as important on the coast as in the river valleys due to very poor nature of the coastal soils. The researchers also agree that there is a low frequency of ceremonial sites with mounds in the coastal environmental zone. The lack of mounds in the area may reflect lower agricultural production potential and consequent lesser support of a chiefly ruling class.

Historic Overview

As can be seen in Table 2, the historic chronology of northwest Florida begins with the Spanish explorers who sailed into the Pensacola Bay system. Pensacola was included in two well known Spanish explorations: Panfilo de Narvaez in 1528 and Hernando de Soto in 1539.
First Spanish Period (1528-1763)

The first Europeans to see present-day northwest Florida were probably members of the Narvaez expedition in 1528 (Hodge 1907:37-40). Cabeza de Vaca's narrative of this expedition produced the first description of the northwest Florida natives. The next known Spanish visitor to Pensacola Bay was the pilot of the Soto expedition, Diego Maldonado, who sailed into the bay during the winters of 1539 and 1540 to resupply the ill-fated expedition. Soto never arrived, choosing to remain inland, and there are no known descriptive documents from the Maldonado visit. During his sojourn in the Pensacola Bay region, Maldonado probably explored the rivers and bayous (Lewis 1907:193).

Europeans returned to the Pensacola area in 1559. Under pressure to establish a foothold in the New World north of Mexico, the Spanish sent Tristán de Luna with a large contingent of colonists to establish a settlement at Pensacola. He was also under orders to move to Coosa (northern Georgia) and St. Helena (Parris Island, South Carolina). Within weeks of landing the 11 to 13 vessel fleet and before the ships were completely unloaded, a hurricane destroyed 7 to 9 of the ships. This setback, in combination with diseases and dissension, doomed the Luna colony and it was abandoned in 1561 (Priestly 1928). The location of the Luna settlement on Pensacola Bay is unknown. Recent underwater investigations in Pensacola Bay by Roger Smith, Underwater Archaeologist for the State of Florida, have revealed the location of a sixteenth century shipwreck that is associated with the Luna expedition (Smith et al. 1995). Subsequent excavations of the shipwreck by Smith and archaeologists John Bratten and J. COZ Cozzi of the UWF maritime archaeology group recovered a large number of Spanish artifacts from the early to mid-sixteenth century and provided details about sixteenth-century ship construction (Smith et al. 1998).

Nearly one hundred and fifty years passed before the Spanish once again became interested in colonizing the Pensacola area. In 1698, fearful of French expansion, the Spanish founded the presidio of Santa Maria de Galve and Fort San Carlos de Austria on Pensacola Bay, and the long Spanish colonial period commenced. The French attacked Santa Maria de Galve in 1719, and burned the outpost. The Spanish and French continued to struggle over the presidio between 1719 and 1722. In 1722, following a treaty between the two warring nations, France restored northwest Florida to the Spanish (Parks 1986). The University of West Florida recently completed a multi-year project to investigate First Pensacola on Pensacola Naval Air Station (Bense and Wilson 1999). This work, combining extensive historical research and large-scale archaeological investigation, isolated and spatially defined the First and Second Spanish and British colonial occupations, as well as a Native American component. The temporally associated First Spanish and American Indian components are represented by sealed midden deposits and features and two cemetery with at least 17 interments. This work also identified the stockade walls of Fort San Carlos de Austria and several structures, numerous features, and thousands of artifacts within this military community.

After the 1722 treaty was signed, the Spanish resettled on Santa Rosa Island. Their new fort, known as Santa Rosa Punta de Siquenza (8ES22), was destroyed by a hurricane in 1752 (Griffen 1959). The survivors moved back to the mainland near a blockhouse and mission in the area of Seville Square in downtown Pensacola. Few colonial settlers ventured past the small outpost of Pensacola on the bay, fearful of the still numerous and unpredictable Indians in the surrounding region, although maps drawn in the 1770s by English surveyor David Taitt show a First Spanish fort along the Escambia River near present day Molino and "spanish cowpens" to the south between Spanish Mill Creek and Williams Creek. The Spanish were aware of the value of the pine and cypress forests of the northwest Florida region. Don Andres de Ariola, future commander of Presidio Santa Maria de Galve, attempted as early as 1695 to harvest from the East River swamps yellow pine and cypress logs for masts and spars. Logging in this area continued intermittently into the early 1700s (Hunter 1999). In 1743, Dominick Serres was shipping masts of yellow pine to Havana, Cuba (Gober 1956:104; Polk 1971:1). By royal decree the settlement was renamed Panzacola. The archaeological remains of the stockade and interior buildings of Fort Panzacola have been identified in the City of Pensacola (Joy 1989b, Bense 1999).

 

British Period (1763-1781)

The British gained control of Florida from the Spanish following the Treaty of Paris in 1763. Pensacola became the capital of the Province of West Florida. They converted the small Spanish settlement into a heavily fortified military establishment, busy port city and center of commerce (Parks 1986). Maps drawn by George Gauld in 1768 and David Taitt in 1771 indicate considerable British activity in the interior, including several large land grants along Escambia and East bays, and a number of huts and small temporary Indian camps along the road leading from Pensacola to the villages of the Upper Creek Nation near Montgomery, Alabama and the confluence of the Coosa and Tallapoosa rivers. Outside Pensacola, the British operated at least three water-powered sawmills. Gauld's 1768 map of the Pensacola area shows Tate's Sawmill on Elevenmile Creek near Perdido Bay. Snider and Palmer (1994:549) note the "Old English Sawmill" north of Pensacola on a tributary of the Escambia River. The Colonial Office Records document the James Bruce sawmill that is probably located on Carpenter's Creek (Phillips 1996; 1998). Timber, indigo, deerskins, cattle, corn, tallow, bear's oil, rice, tobacco, salted fish, pecans, sassafras, and oranges were exported during this period (Howard 1940:217). The archaeological remains of the Fort of Pensacola and interior buildings, private residences outside the fort, and the nearby Fort George redoubt, have been documented in several investigations (Baker 1975; Bense 1989; Cusick 1986; Joy 1989b, Bense 1994a; 1999).

By the late 1770s the English and Spanish were again at war. Spanish forces under Bernardo de Galvez, attempting to destroy British influence on the northern Gulf Coast, ousted the English from Pensacola in 1781 following the Siege of Pensacola (Coker and Coker1981). This battle, fought in the North Hill area was archaeologically documented by Baker (1975). The Spanish regained control of northwest Florida, and established Fort San Carlos de Barrancas on the mainland at the mouth of Pensacola Bay to protect the harbor.

Second Spanish Period (1781-1821)

During the Second Spanish Colonial period (1781-1821) the local population continued to grow. Indian trade grew in commercial importance, more brickyards and sawmills were established, and cattle ranching thrived. The vast longleaf pine forests of northwest Florida became even more important economically, and northwest Florida's considerable topographic relief and many spring-fed, perennial drainages provided countless water-powered mill seats for the lumber industry. Although the pine forests were exploited during the British period, the Second Spanish period witnessed an increase in milling activity as timber became an even more important resource. In 1798 Milan de la Carrera constructed a sawmill near the Escambia River (American State Papers 1859:173; Yonge 1909:72) and later built a second sawmill nearby (Snider and Palmer:549). One of these mill sites (8ES1965) is located on Clear Creek (cf. 1815 Pintado map on file at UWF Special Collections Library), and the second sawmill site (8ES982) is located to the north on Spanish Mill Creek. By 1815, Thomas English operated a sawmill near Perdido Bay on Elevenmile Creek (English's Sawmill Creek), and in 1816 Joseph Phillips erected a sawmill (8ES1368) north of Pensacola on Indian Pass (Carpenter) Creek (Snider and Palmer 1994). Also during the Second Spanish period, Jose Noriega may have operated a sawmill on Graveyard Branch (Noriega's Mill Creek or Sawmill Branch), and to the north of Milan de la Carrera's operation, a sawmill probably operated on Big Rock Creek (8ES31). The massive Arcadia mill complex (8SR384) is another of these water-powered industries that had its origins late in the Colonial period. The Arcadia property was given as a Spanish land grant to Juan de la Rua, a prominent Spanish businessman in Pensacola, during the waning years of the Second Spanish period (ca. 1817). The Arcadia mill was fully developed during the following Early American period into a productive industrial complex that dwarfed all other water-powered mills in the area (Phillips 1993a).

The Second Spanish Period also saw the rise of a profitable market for deerskins. The Industrial Revolution in England during the late 1700s created a strong demand for leather for belts and bands, and leather from West Florida deerskins became an important trade commodity. Northwest Florida's chief commercial export became deer hides, and the Panton-Leslie trading company, with its main office at Pensacola, grew into a thriving firm which traded blankets, guns, and other articles with the southeastern Indians in exchange for deerskins. Begun in 1783, Panton, Leslie and Company came to dominate the Indian trade of the region. Pensacola became the center of a large trading operation which netted the company significant profits from deerskins and eventually gained them vast Indian lands (Brown 1959:328-336).

By 1800, however, Pensacola and the northwest Florida region were experiencing severe economic and political difficulties that were related to international developments. Pensacola was a poor, undermanned military outpost, and Spain's slipping political, economic, and military strength meant continuous difficulties for the small colonial port city. These international developments alarmed Spain. They viewed with apprehension Napoleon's rise to power, French imperialistic designs on the New World, and the sale of Louisiana to the United States. Also contributing to Spain's and Pensacola's difficulties were an increasing population of aggressive Anglo-Americans on the Spanish Florida borderlands; the inevitable Anglo-American and Indian skirmishes; and the growing conflicts between Great Britain and the United States. Pensacola was vulnerable to attack and expropriation in this volatile political environment. Starting with the Creek War and the War of 1812, and the accompanying British and Indian intrigues along the American border, northwest Florida became a target for American military predation. Twice, in 1814 and again in 1818, General Andrew Jackson led U.S. forces into Spanish Pensacola to restore order along the frontier. These maneuvers convinced Spain that Florida was a liability, and the Spanish colony was officially transferred to the United States in July of 1821 (Parks 1986:43-48; Rucker 1990:6-35). Archaeological investigations in the City of Pensacola (cf. Bense 1989; 1999) and in southern Escambia County have documented a wide variety of settlements and features associated with the Second Spanish Period.
Many of the colonial settlements were concentrated near the mouth of Pensacola Bay on the peninsulas, islands, and mainland of present day Escambia County. However, several areas of coastal Santa Rosa County were intensively utilized during the Colonial period. Among the most significant is the ship careening area on the western tip of the Gulf Breeze peninsula. Here, along the shores of Old Navy Cove, English Navy Cove and Town Point, ships were cleaned and repaired throughout the Colonial period. Part of this area has been surveyed (Joy 1988), and a scuttled shipwreck from the British period has been identified and excavated by the University of West Florida (Bense 1988; Franklin, Morris, and Smith 1992). Many more shipwrecks lie just off the western end of the Gulf Breeze peninsula.

Anglo-American settlers from the Carolinas also came to the northwest Florida interior late in the Second Spanish Colonial period looking for better opportunities. They settled in mixed pine-oak forests along small drainages near the Escambia and Blackwater rivers on marginally productive agricultural soils, and in the upland drainage divides near present-day Jay and Walnut Hill in rich agricultural soils. These settlers cleared the forests, established small logging and farming communities, and constructed water-powered sawmills and gristmills. These settlers essentially "squatted" in what was Spanish territory. The Spanish contingent at Pensacola, however, did not mount any serious efforts to remove these settlers, possibly because the Anglo-Americans provided much needed farm produce to the inhabitants of the city (Wilson 1999). The archaeological remains of one these early Anglo-American communities has been found on the terraces along Diamond Creek, a tributary of the Escambia River on the western side of Santa Rosa County (Bense 1983). Many more of these late Second Spanish and Early American settlements exist, but have not been archaeologically documented.

American Period (1821-present)

Florida became an American Territory in 1821 after a series of successful invasions of Pensacola by Andrew Jackson in pursuit of American Indians. These Creek Indians were hostile to American expansion and had sought refuge in Pensacola. The Spanish finally sold Florida to the United States and withdrew, ceding Florida by treaty to the United States in 1819. King Ferdinand of Spain signed the treaty in 1820, and it became an American territory in 1821 (McGovern 1974). Pensacola was the temporary capital of this new territory and Jackson became interim governor.

Between 1821 and 1861, the region grew slowly. Pensacola's economic condition improved in the 1820s when a U.S. Navy Yard was established southwest of the city; however, Pensacola's development suffered from periodic yellow fever epidemics, government neglect, and the lack of a viable hinterland. Nonetheless, the forests and bays of northwest Florida offered economic opportunities. Numerous brickyards in the vicinity of Pensacola and along the larger rivers in northwest Florida, provided bricks for the federal forts under construction near the mouth of the harbor: Pickens (1834), McRea (1840), and Barrancas (1844). These brickyards also supplied bricks to other Gulf Coast locations. More importantly, the lumber industry grew even larger and more economically important during this period, and the huge stands of longleaf yellow pine were soon being harvested throughout northwest Florida. More water-powered sawmills and grist mills sprang up on the streams of present-day Escambia and Santa Rosa counties, giving birth to additional small frontier communities. Among these Early American water-powered industries were McVoy's sawmill (8ES1965) on Clear Creek just north of the University of West Florida campus, Loften Cotton's Mill (8SR736) on Clear Creek northeast of Milton, and Keyser's sawmill (8SR930) in Milton. These and dozens of other small water-powered mill sites have been investigated by UWF archaeologists since 1989 (cf. Phillips 1989a, 1993b, 1996; Penton 1991).

In the 1840s, steam-powered lumber mills began to appear along the rivers and bays. As steam engines required no moving water to power the saws, this new technology freed the big lumber operations from the dual constraints imposed by the traditional interior-based, water-powered technology of poor transportation networks and an uneven and limited power supply. The new technology enabled the lumbermen to move down out of the interior, closer to the commercial shipping lanes. As the center of the lumber shifted from the interior to the lower rivers and bays; larger mill towns such as Milton, Bagdad and Pinewood (Bay Point Mill Company) began to develop and prosper; Pensacola also had a thriving lumber industry on Point Diablo (8ES1361). Additional noteworthy antebellum steam powered mill complexes include Stearns sawmill (8SR390) near Mulat on Escambia Bay, and the sawmill at Miller's Point on East Bay, both in Santa Rosa County. By the first days of the Civil War, Pensacola finally achieved its first railroad connection to Montgomery, Alabama, and the future appeared promising (Dibble 1974; Doherty 1959:337-356; Parks 1986:49-66).

When the Civil War began in 1861, Floridians who lived in the western panhandle area had mixed loyalties. Some citizens from Pensacola, Marianna, and Milton hoped to delay secession or postpone it indefinitely, whereas most northwest Floridians were anxious to sever ties with the North (Adams et al.1992). The federal garrison at Fort Pickens, located at the entrance to Pensacola Bay, refused to surrender and was one of the few Southern fortifications held by the Union for the duration of the war. The Confederates unsuccessfully attacked Fort Pickens in October 1861, and artillery bombardments took place in 1861-62 between Pickens and nearby Confederate-held forts. But by early 1862, with more strategic regions of the South in peril from Union advances, Pensacola was abandoned by the Confederacy. After implementing a scorched earth policy on the region's industrial complexes, the Confederate forces retreated from Pensacola in the spring of 1862. Pensacola was practically abandoned for the remainder of the war, and various skirmishes between Union forces and Confederate forces occurred throughout northwest Florida until 1865 and the war's end (Parks, Rick and Simons 1978; Parks 1986: 67-74; Rucker 1990: 625-750). The remoteness of the western panhandle provided a haven for people coming to avoid conscription into the Confederate Army. Those people, some of whom were Union collaborators, supplied Union ground forces and blockade ships with valuable information and guided Union forces on raids throughout the region.

Most citizens of the state welcomed the cessation of hostilities and the opportunity to return to a normal life. The economy, however, was in shambles and property values plummeted (Adams et al. 1992). The lack of adequate transportation to inland areas impeded economic development and population growth. The end of the war also brought anarchy to northwest Florida. Bands of former soldiers, deserters, and criminals terrorized the population. Local governments collapsed and in 1866 several northwest Florida counties were placed under martial law.
Much of northwest Florida's industrial capacity had been destroyed in the war, and the region laboriously began rebuilding. The lumber and turpentine industries enjoyed a boom in the late 1800s. Pensacola became one of the world's best known ports for the exportation of yellow pine timber, shipping it as far away as South America, Italy, and Scandinavia. Lumber railroads stretched deeper into the forests and tapped previously inaccessible stands of pine. These railroads provided transportation for the lumber industry and greatly facilitated logging in the interior uplands away from the drainages. As the interior ridges were logged, farming communities began to develop on the agriculturally rich sandy loams found in these uplands. Meanwhile, scores of large mills sprang up along the shores of Perdido, Escambia, East, and Blackwater bays. Among these steam powered lumber mills were the Bluff Springs, Milner, McDavid, and McMillan operations in north Escambia County, Millview on Perdido Bay, and a number of mill complexes in Pensacola ( Phillips 1993b). The Bagdad and Bay Point mills on Blackwater Bay continued to thrive. In about 1901 the massive Alger-Sullivan Mill began operation and company town of Century was born in extreme northeast Escambia County (Fischer and Collier 1993:79). Smaller lumber mills, boat repair yards, boat works and ports also developed along the rivers and smaller drainages. Lumbering and turpentining communities and temporary camps were set up throughout the interior of the area. Railroad connections were re-established to the north, and new rail lines connected Pensacola east to Jacksonville and west to other Gulf Coast communities. A growing fishing industry, primarily exploiting red snapper, also developed in the Pensacola vicinity by the late 1800s ( Parks 1982; McGovern 1976; Parks 1986:85-109; Pearce 1980: 90-179; Rucker 1990: 758-761; Wells 1976).

Unfortunately, the failure of northwest Florida's lumbermen to implement resource conservation measures brought the lumber boom to an end by the early decades of the twentieth century. The Bay Point mill fell silent in the early 1920s, and the mammoth Bagdad Mills ceased operation in 1939; only the Alger-Sullivan Lumber Co. in Century survived. Increasingly the inhabitants of Florida's panhandle turned to agriculture, planting the cut-over timberlands in cotton, corn, and peanuts, and experimenting with blueberries, satsumas, pecans, grapes, and pears.

The Great Depression further weakened an already stagnant economy. The onset of World War II improved conditions, however, and in the years following the war, the presence of the U.S. military profoundly altered and improved the economic climate. Military installations like Pensacola Naval Air Station, Whiting Field, and Eglin Air Force Base have proved extremely important in the development of northwest Florida since the mid-twentieth century. Agricultural pursuits and the pulpwood industry are still dominant in the interior portions of northwest Florida, but the economy has diversified in the last thirty years with the addition of chemical plants, oil fields, and beach-related tourism to the local economy (Armstrong 1930; Fischer and Collier 1993; King 1972; McGovern 1976; Parks 1986:85-109; Pearce 1980:90-179; Rucker 1990:758-761; Wells 1976).