Three cultural resources have been located on the property; two prehistoric
archaeological sites and one underwater historic shipwreck. The underwater
site was known prior to the C.A.R.L. survey of the property, but the two
prehistoric sites were previously unknown. Research was also conducted
on the historic towns of Old and New Troy.
Beginning around 12,000 years ago, sites identified as belonging to
the PaleoIndian tradition appear in Florida. These sites are marked
by the presence of particular types of stone tools--most commonly Suwannee
and Simpson projectile points, unifacial and bifacial blades and knives.
Because of lower sea levels, groundwater levels in the central upland regions
of the state were much lower and the landmass of Florida was about twice
what is today. Many of the known PaleoIndian sites are associated
with water sources such as deep springs, shallow lakes or prairies with
associated limestone strata (Milanich 1994:2).
Beginning around 9,000 B.C. water levels began to rise reducing the size of Florida. By around 7,500 B.C. environmental changes had sufficiently altered the lifestyles of inhabitants for a separate period to be recognized by archaeologists. The Archaic period is marked by an increase in population, a more settled lifestyle, and a larger variety of specialized tools. By around 2,000 B.C. pottery was in use, and by the end of the late Archaic period (500 B.C.) regional cultures are recognized.
As mentioned, Troy Springs is located at the juncture of three culture areas. By around 500 B.C. several separate manifestations of the Weeden Island culture were apparent in the region. The early Weeden Island ceramic assemblage is characterized by the presence of Swift Creek complicated-stamped pottery, punctated and incised Weeden Island types, and undecorated pottery. By later Weeden-Island times, the Swift Creek pottery disappears and is replaced by Wakulla Check Stamped pottery. Weeden Island is notable because of the distinction between secular and sacred pottery styles, with mortuary and ceremonial contexts containing animal and anthropomorphic effigy vessels.
By around A.D. 750 sufficient changes had occurred in the ceramic inventory for archaeologists to recognizes two new cultures in the region, the Suwannee Valley and Alachua cultures. These cultures appear to have developed out of, and perhaps overlapped, the late Weeden Island cultures of the area. The Suwannee Valley cultural assemblage consists of undecorated and decorated pottery. Types of decoration consist of surface-roughening, check-stamping, cord-marking, cob-marking, punctating, and brushing. The Alachua cultural assemblage is very similar, but is found slightly to the south of the Suwannee Valley region and exhibits differences in the proportions of cob-marked pottery (Milanich 1994:348-353; Johnson and Nelson 1990:48-61; Worth in Weisman 1992:188-205). Indian Pond, a more specific ceramic tradition within the Suwannee Valley culture, is recognized by the presence of cord and fabric-marked ceramics and a poorly understood linear-marked pottery type (Johnson and Nelson:1990:60).
Just before, during, and after contact with Europeans a cultural complex identified as Leon-Jefferson is recognized. Changes in surface decoration, rim modification, vessel form, and ceramic technology, as well as other cultural changes are associated with Leon-Jefferson. The addition of European goods and technology to the existing prehistoric lifestyles produced profound changes in the aboriginal populations.
Following the Spanish conquest of Florida a system of missions was established
across North Florida and the panhandle. The Spanish missions were
designed to control native populations, supply a labor source, and create
a defensive line against the English. A number of mission sites have
been identified and excavated by archaeologists in North Florida (Milanich
1995:167-183). Disease and English attacks on the mission Indians
caused massive depopulation of the North Florida region. The absence
of native populations combined with uneasy English-Indian relations lead
to the immigration of Creek and related tribes in the mid-18th century.
These Indians eventually became known as Seminoles.
Florida was ceded from Spain to the United States in 1821. At this time the Seminoles had become prosperous in the North and North Central Florida regions, raising cattle as the Spanish had done. American politics, pressure from settlers, and uneasiness towards the Indians lead to the 2nd and 3rd Seminole Wars, intended to subdue and remove the Seminoles from their Florida lands to Oklahoma.
The town of Troy, or Old Troy, "located about 8 or 9 miles up the Suwannee
River from present-day Branford," served as the first county seat of Lafayette
County. The exact date this community was settled is unclear, though
Melton (1974) notes that a post office was established at Old Troy on 27
July, 1845. Postal records indicate that a post office was established
at Troy (spelled "Tray" in the official records) on 22 July 1846 and discontinued
31 July, 1847 (Pickett et al. 1957:50). Some sources say the county
courthouse, as well as other structures, burned in 1865, near the close
of the Civil War, by Union sympathizers and deserters from the Confederate
Army (Warnke 1971:30; undated article from the Gainesville Sun).
Lafayette and neighboring counties were a haven for deserters and Union
sympathizers; in 1864 Major Campfield and 300 Confederate calvary and 225
infantry swept through the region hunting deserters and burned the homes
of many loyal citizens as well (Melton 1974). Melton (1974) reports
that Old Troy burned in 1856, well before the beginning of the Civil War.
The exact location of Old Troy remains a mystery. Melton (1974) states that the settlement was built on the banks of Troy Springs. This, however, was not an ideal spot, since river flooding would fill the nearby sloughs and cut the settlement off from the river and nearby trails. Melton (1974) further states that the remains of old log houses were seen at Troy Springs, though no evidence remains today. When the town of Old Troy was burned it was comprised of a one-story log courthouse, five stores, two doctors, a saloon, and post office. Legend suggests that the county judge was warned of the impending disaster and was able to remove important papers from the town.
The survey plat of Township 5 South, Range 13 East, dated 1852, was
examined for evidence of the location of Old Troy. No indication
of the town was found; Troy Springs is simply labeled "lagoon." Some
subdividing of sections around Troy Springs and New Troy is shown, perhaps
providing us with some indication of settlement in this general area.
New Troy had its share of problems following the end of the Civil War.
Reconstruction attitudes and a high number of Union sympathizers in the
county were confronted with increasing activity by the Ku-Klux-Klan.
Friction between Republican and Democrat heightened the tension.
Bell (1994) reports on the assassination of Judge John Krimminger in New
Troy in 1871. Krimminger was active in local and state politics,
and as a Republican and Confederate deserter was not well liked.
Bell (1994) notes that a number of murders were committed, and there was
a general consensus that the Klan was involved. Despite a number
of witnesses, as well as the testimony of Krimminger's widow, the assassin
managed to evade capture by moving from county to county, and by relying
on sympathetic Democrat friends.
A rather interesting account of New Troy in the 1880s is presented by
Eugene W. Sears in Romantic History of New Troy, Capital of Lafayette County,
written in 1936 as part of the Federal Writers Program. Sears was
apparently the descendant of one of New Troy's pioneers, and used their
family stories and recollections as the basis for his essay. Bell
(1990) has recently republished this fictionalized essay with some notes
on the characters and settings. Sears (in Bell 1990) describes the
community as a thriving one, with a grist mill; two general stores; hotels;
log houses with stick and clay chimneys; a Baptist Church with sermons
by circuit riding preachers of various denominations; and, of course, the
two-story courthouse. New Troy was also serviced by a ferry.
An undated newspaper article on file at the Lafayette County courthouse
lists other features of New Troy in the 1890s, including a cotton gin,
sawmill, jail, boarding house, as well as Baptist and Methodist churches.
Two newspapers were produced in New Troy-- The Lafayette County Messenger
and The Lafayette County Enterprise, though neither is preserved in library
collections today. The area was also a hub for steamboat traffic,
and products like turpentine, cotton, vegetables, and oranges were shipped
out.
The New Troy courthouse caught fire and burned down New Year's Eve, 1892. Shortly after this county residents voted to move the county seat to Mayo, where it is today. The removal of the courthouse spelled the end for the little community of New Troy, and business soon began to fade. The undated article from the Gainesville Sun states that houses were dismantled for their timber and bricks, hardwoods replaced the fields, steamboat traffic ended in 1899, and the ferry closed in 1917.
During the months of May, June, and July, twenty-one 50x50 centimeter
shovel tests were dug on the property. All but three of the tests
were tied into one of two datums and a grid system was used for the placement
of the tests. The first datum was an electric pole and the second
datum was a telephone riser. The remaining three shovel tests (numbers
19-21) were placed arbitrarily. Two sites (8LF55 and 8LF56) were
identified as a result of the C.A.R.L. investigations.
Site 8LF55 was found during shovel testing. Little surface exposure
was noted at the site. Sixteen shovel tests were excavated to determine
the boundaries, both vertically and horizontally, and the cultural affiliation
of the site.
Site 8LF56 was first identified through surface inspection. A scattering of lithic material was noted on the surface. Five shovel tests were placed in the area to determine cultural affiliation, size, and depth.
Through the years all but the hull has been salvaged. A detailed
study of the remains of the Madison is being conducted by graduate student
Richard Haiduven of East Carolina University and should be forthcoming.
Troy Springs 1 - 8LF55: The site was found during the C.A.R.L.
testing of the property. 8LF55 measures approximately 245 meters
by 90 meters and extends to a depth of one meter below present ground surface
in some areas. The stratigraphy of the soil varies slightly, but
in general the area consists of 0 to 18 cm of grey/brown humic soil, 18
to 45 cm of brown and tan mottled sandy soil, and 45 cm to bottom of mottled
tan and light tan to an orange/tan clay.
Artifacts were recovered throughout all stratigraphic levels at the
site and continued to be recovered at depths of one meter below surface.
Often pottery recovered from such depths is explained as a mixing of upper
levels with lower levels, a common problem in shovel testing, but at the
Troy Springs 1 site, this occurred too often to be explained in such a
way.
Artifacts recovered during the testing include lithics, prehistoric
pottery, bone, modern debris, and assorted material such as pebbles and
iron concretions (see appendix). The majority of the artifacts were
chert--258 pieces out of a total of 310 total artifacts.
Analysis of the distribution of chert artifacts indicates that the heaviest
concentration of material occurs in a roughly oblong area defined by shovel
tests 1, 2, 3, 4, 6, and 12 (note that tests 1, 2, 3, 12 produced the majority
of the pottery at the site). Of 258 chert artifacts, 201 (78%) came
from these 6 tests. Analysis of the chert artifacts from this part
of the site produces the following pattern: 140 (70%) complete flakes;
26 (13%) broken flakes; 26 (13%) fragmentary flakes; 7 (3.5%) debris; and
1 (.5%) core. In examining the complete flakes they were found to
be larger than the broken and fragmentary flakes and they also retained
cortex in 12 (8.5%) cases. The pattern observed here does not closely
resemble the behavioral/technological groups defined in Sullivan and Rozen
(1985:762-763). However, the high number of complete flakes would
suggest core reduction, rather than tool production, as suggested by the
Sullivan and Rozen (1985) model.
Thirty-three pieces of pottery were recovered from five shovel tests
at the site. Depths ranged from just below the surface to 100 cm
below surface. The majority of the recovered pottery was sand-tempered
plain (78%)(Table 1). Assignment of this type to a specific period
is difficult as it appears in many pottery-producing cultures. Sand-tempered
check-stamped pottery (12% of total recovered) also is difficult to assign
as it is common in several time periods. However, the linear-marked
sherds (6% total) are representative of Indian Pond, a ceramic complex
associated with the Utina Indians of North Florida. They occupied
a territory lying roughly between the Santa Fe and Suwannee Rivers, extending
northward into Georgia (Johnson and Nelson 1990:48). While commonly
associated with the early 16th century, an early manifestation of Indian
Pond (between A.D. 950 and 1250) is present at a site adjacent to the Ichetucknee
River (Milanich 1994:349). One Weeden Island plain sherd was recovered
from the 8LF55
. The Weeden Island period predates, and during the later periods
was probably contemporaneous with, Indian Pond. Taken as a whole,
the ceramic assemblage can be viewed as representative of early Indian
Pond culture. Although many pottery types usually associated with
Indian Pond, such as fabric and cord-marked pottery, were not recovered,
the total assemblage supports assignment to this culture. The general location
of some artifacts was noted during excavation of the shovel tests.
The majority of the material recovered often came from the second zone
of the shovel test, described as a brown/tan mottled soil in units 1, 2,
3, and 4. In many of the deeper units, sherds and chert flakes were
recovered from the deepest zones, directly above the orange clay that underlies
much of the area. This includes the Weeden Island Plain sherd, some
of the sand-tempered check stamped sherds, as well as sand-tempered plain
sherds.
Troy Springs 2 - 8LF56: This site was discovered during C.A.R.L. testing
of the property. The site is approximately 96 meters by 90 meters,
extending from the surface to a depth of approximately 50 cm. Material
found on the surface and in subsurface tests is limited to undiagnostic
chert flakes and no more than 5 flakes were found in a given shovel test
(see appendix). Despite the sparse nature of the artifactual material,
flakes were recovered from depths of 15 to 70 cmbs. A decorative
brass(?) fitting was recovered in one shovel test to the north of 8LF56,
and may be associated with the historic occupation of the Troy Springs
area.
Most of the planned recreational activities on the property, such as
boating, swimming, and picnicking, will take place in and around the spring.
Two of the sites are located in this area, 8LF5-The Madison and 8LF55-Troy
Springs 1, and it is likely that they will be impacted as a result of these
activities. Careful planning in consultation with the DHR and the
C.A.R.L. Archaeological Survey should be implemented to protect them.
It is likely that a program of restoration of the banks of the
spring will be considered. Again, consultation with DHR is necessary
for protection of the neighboring cultural sites.
Whenever possible the interpretation of cultural resources is encouraged. The setting at Troy Springs provides an excellent opportunity for the Office of Greenways and Trails to tell visitors about the prehistoric and historic occupation of the region. The C.A.R.L. Archaeological Survey can provide support and offer funding sources for an interpretative program at the site. Personnel are encouraged to call either 850/487-2299 or 850/829-9100.
Barker-Benfield, Simon
1995 A Wellspring of Civil War Nautical History. The Florida
Times Union, Jacksonville, Florida.
September 17, 1995
Bell, Wilburn
1994 Murder in New Troy: The Assassination of Judge John
Newton Krimminger, Based on the Testimony of Rebecca U. Krimminger, November
13, 1871. Printed by the author, Day, Florida.
Bradbury, Alford G., and E. Story Hallock
1962 A Chronology of Florida Post Offices. Florida Federation
of Stamp Clubs, Palm Beach.
Caldwell, John M.
1937 The Steamboat Madison. Typescript of an article originally
published in a 1903 Tallahassee newspaper. W.P.A. Federal Writer's Program.
Chance, Marsha A.
1992 An Assessment of Known Cultural Resources on Suwannee
River Water Management District Properties in North Central Florida. Environmental
Services, Inc., Jacksonville.
Fryman, Mildred
1971 Notes on Lafayette County, accompanying site file
for 8LF12, Florida Site File, Bureau of Archaeological Research,
Tallahassee.
Johnson, Kenneth W. and Bruce C. Nelson
1990 The Utina: Seriations and Chronology. The Florida
Anthropologist
43:48-62.
Melton, Holmes, Jr.
1974 Lafayette County History and Heritage: An Anthology
[Collection of newspaper articles, primarily by Melton].
Milanich, Jerald T.
1994 Archaeology of Precolumbian Florida. University Press
of Florida, Gainesville.
1995 Florida Indians and the Invasion from Europe. University Press of Florida, Gainesville.
Milanich, Jerald T., and Charles H. Fairbanks
1980 Florida Archaeology. Academic Press, New York.
Office of Environmental Services, Division of State Lands, Department
of Environmental Protection.
1995 Conservation and Recreation Lands Annual Report 1995.
Board of Trustees of the Internal Improvement Trust Fund. In cooperation
with Land Acquisition Advisory Council, Tallahassee.
Pickett, Lee N., Kenneth L. Rice, and Henry M. Spelman III
1957 Florida Postal History and Postal Markings During
the Stampless Period. Florida Federation of Stamp Clubs, Palm Beach.
Rosenau, Jack C., Glen L. Faulkner, Charles W. Hendry, Jr., and Robert
W. Hull
1978 Springs of Florida. Bulletin 31 (revised), Bureau
of Geology, Tallahassee.
Sears, Eugene W.
1936 Romantic History of New Troy, Capital of Lafayette
County, During the Early Eighties, edited and republished by Wilburn
Bell [1990], Day, Florida.
Sullivan, Alan P. III, and Kenneth C. Rozen
1985 Debitage Analysis and Archaeological
Interpretation. American Antiquity 50(4):755-779.
Weisman, Brent R.
1992 Excavations on the Franciscan Frontier, Archaeology
of the Fig Springs Mission. University Press of Florida, Gainesville.