78 THE FLORIDA ANTHROPOLOGIST 2004 VOL. 57(1-2)
Table 5. Radiocarbon dates from FS 1031 (bottom half of Feature 690 infilling).
Sample Data Measured BCPC Conventional
Radiocarbon Age Ratio Radiocarbon Age
Beta - 169797 1500 + 110 BP -24.6 o/oo 1510+ 110 BP
SAMPLE: DA12 FS1031-SILT
ANALYSIS: Radiometric-Standard delivery (with extended counting)
MATERIAL/PRETREATMENT: (organic sediment): acid/alkali/acid
2 SIGMA CALIBRATION: Cal AD 330 to 700 (Cal BP 1620 to 1250)
Beta - 171116 2310 + 40 BP
SAMPLE: DA12 FS1031-CHARCOAL
ANALYSIS: AMS-Standard delivery
-26.9 0/00
2280 + 40 BP
MATERIAL/PRETREATMENT: (charred material): acid/alkali/acid
2 SIGMA CALIBRATION: Cal BC 400 to 350 (Cal BP 2350 to 2300) AND
Cal BC 310 to 210 (Cal BP 2260 to 2160)
midden. Two major groupings were identified—one lacking
very coarse to coarse sand size fractions, a second that has a
substantial percentage of these fractions, with one sample
transitional between the two. This suggests the possibility for
two depositional events, but also points to the relative unifor-
mity of the deposits across the site.
The study was continued by comparing the size fractions
using a multivariate statistical method involving hierarchical
linkage of the size fractions by particle types. This analysis
shows that samples 1036 and 1033 were most similar to one
another, followed by 1035, 1030, and 1034, which are next
linked to 1037, 313, and 379, and finally linked to 1031 and
329, which are the least similar of ail. This analysis shows
little correlation with the initial groupings. To address this
problem, the study focused on four new data sets dealing solely
with the sand-size fractions of the ten samples. This analysis
demonstrated a more comprehensible pattern based on one size
fraction (+1 to +2 on the Phi grain size scale); the result is that
samples 1031, 1034, and 1035 are most similar, followed by
samples 1030, 1036 and 1037. Samples 313 and 379 form a
third group, with samples 329 and 1033 being dissimilar to all
three groups and from one another. There is some correlation
between the particle size groups initially identified and the
hierarchical cluster analysis groups, though it is not possible
at this point to explain the differences between the three
groups and the two outlier samples (320 and 1033).
One soil sample of the very large FS 1031 (bottom half of
Feature #690 infilling) was selected for bulk radiocarbon
dating. The standard Beta Analytic, Inc. pretreatment process
produced two fractions—about 8.0 g of black silt plus 1.5 g of
identifiable particulate (sand-size) charcoal. They were dated
separately, the silt by conventional radiocarbon analysis (with
extended counting) and the charcoal by AMS. The conven-
tional ages of the two dates differ by 770 years and the 2 sigma
calibrated ranges do not overlap. The difference in age
between the silt and charcoal fractions has important implica-
tions for understanding the depositional history of the site—it
may represent two depositional episodes or be related to
downward migration of the silt fraction within the soil column.
Directions for future research include focusing on the
medium sand fraction in a larger number of samples, addi-
tional bulk radiocarbon dating of the soils, as well as tests such
as % organics, pH, chemical analysis, magnetic susceptibility,
and micromorphology.
Acknowledgments
This study would not have been possible without the interest of
Bob Carr in pursuing new pathways to understanding South Florida’s
prehistoric midden deposits. For her careful sedimentological
analyses of these samples I thank Ms. Tina Manne, a graduate
student in the Division of Marine Geology and Geophysics of the
Rosenstiel School of Marine and Atmospheric Science. Lastly, I
thank Dr. Matt Lynn, Associate Director of the University of
Miami’s Center for Advanced Microscopy.
References Cited
Doran, J.E., and F.R. Hodson
1975 Mathematics and Computers in Archaeology. Harvard
University Press, Cambridge.
Hoffmeister, J.E
1974 Land from the Sea: the Geologic Story of South Florida.
University of Miami Press, Coral Gables.
Hoffmeister, J.E., K.W. Stockman, and H.G. Multer
1967 Miami limestone of Florida and its Bahamian counterpart.
Geological Society of American Bulletin 78: 175-190.
Masson, M., R.S. Carr, and D.S. Goldman
1988 The Taylor’s Head site (8Bd74): sampling a prehistoric
midden on an Everglade tree island. The Florida Anthro-
pologist 41(3): 336-350.