Kise GEOCHEMICAL STUDIES OF PUMICE 237 identical with the trachyandesite from Wynnhaven Beach (Figure 16b). If the mafic pumice from Florida were derived from the eastern Mexico volcanic provinces, transport would have been by current systems within the Gulf of Mexico with an overall transport distance of 2000 kilometers. The most likely route of transport for pumice from the volcanic islands off the coast of Africa to southeastern Florida would involve transport via the Atlantic Equatorial Current into the Gulf of Mexico and then transport by the Florida Current into the main Gulf Stream near the Florida Straits (Figure 21b), a total distance of over 5000 km. Summary and Conclusions Most of the pumice artifacts from the Miami Circle-Brickell Point archaeological site (8DA12) and other sites in southern Florida are composed of potassium-rich, calc-alkaline rhyolitic material. The pumice was probably transported by ocean currents and deposited on beaches near the sites. Pumice artifacts found at inland sites were probably introduced by local exchange with settlements located near the coastline. A limited number of other types of pumice also have been identified, including rhyolite with lower potassium content, andesite, and pumice that has an alkaline basalt composition. Each type of pumice probably originated from a separate volcanic source. The high silica, high-potassium rhyolitic pumice, which is the most common type of material found in southern Florida, does not match known volcanic material found in the relatively nearby Lesser Antilles volcanic arc. Rhyolitic pumice beds that are present in the easternmost portion of the Trans Mexico Volcanic Belt are chemically similar to the rhyolitic pumice artifacts found in Florida. However, at present, no specific source location can be established within the area. The pumice units present in Mexico are older than the age of archaeological sites in Florida. If this material was transported to Florida the processes would be multistage, involving erosion, river trans- port, and the ocean current transport to the Florida shoreline. The limited number of rhyolitic, dacitic, and andesitic pumice clasts that are characterized by having lower potassium concen- trations could have originated in the Lesser Antilles, perhaps by direct eruption events. This pumice also have traveled within the Gulf of Mexico to reach the southeast Florida shoreline. The source of the two mafic samples analyzed in this study is problematic. Both samples have elevated concentrations of Nb, Ta and Ti, a chemical signature not associated with basalts from island arc regions such as the Lesser Antilles. Alkaline basaltic rocks in eastern Mexico do not have all of these characteristic, however, most of the alkaline basalts of the Cape Verde and Canary Islands do have these trace element signa- tures. However, all the volcanic islands of the eastern Atlantic also have extensive exposures of pumaceous, felsic phonolite and trachyte. Material of this composition has not been observed in the collection of pumice material from Florida. Perhaps an unusually violent Plinean-type eruption was responsible for distribution of enough mafic pumice into the ocean so that some material reached the shores of the western Atlantic, or perhaps the physical characteristics of felsic pumice, such as bulk density and overall porosity, prevented the material from remaining afloat long enough to reach locations in Florida. Notes Surface current trajectories derived from all NOAA AOML Drifting Buoy Assembly data are available at the following web site: http://oceancurrents.rsmas.miami.edu. The drifter “spaghetti map”for the Florida region displays the lack of major currents southeast of Florida (in the northern portion of the Carribean, and the local near- shore perturbation of the Florida Loop Current near the western shoreline of Florida. Acknowledgments I would like to thank Ryan J. Wheeler for introducing me to the problem of pumice occurrences at Brickell Point and other sites in Florida.. Ryan also helped in obtaining all samples used for analy- ses in this study. Greg Peters and Lori Hammer helped with the initial studies of the pumice as part of the Summer 2002 session of the Florida State University Young Scholars Program. References Cited Arculus, R.J. 1976 Geology and Geochemistry of the Alkali Basalt-Andesite Association of Grenada, Lesser Antilles Island Arc. Geo- logical Society of America Bulletin 87:612-624. Baker, P.E. 1984 Geochemical Evolution of St. Kitts and Montserrat, Lesser Antilles. Journal Geological Society of London 141:401-411. Binns, R.E. 1972 Composition and Derivation of Pumice on Postglacial Strandlines in Northern Europe and the Western Arctic. Geological Society of America Bulletin 83:2303-2324. Blake, W., Jr. 1970 Studies of glacial history in Arctic Canada. I. Pumice, Radiocarbon Dates, and Differential Postglacial Uplift in the Eastern Queen Elizabeth Islands. Canadian Journal of Earth Sciences 7:634-664. Boynton, W.V. 1984 Geochemistry of the Rare Earth Elements: Meteorite Studies. In Rare Earth Element Geochemistry, edited by P. Henderson, pp 63-114, Elsevier. Brown, G.M., and J.G. Holland 1977 Geochemistry of the Lesser Antilles Volcanic island Arc. Geochemica Cosmochemica Acta 41:785-801. Bryan, W.B. 1972 Cummingtonite-Dacite Drift Pumice Stranded at West Palm Beach, Florida. Geological Society of America Bulletin 83:3745-3746. Carr, R.S. and J. Ricisak 2000 Preliminary Report of Salvage Archaeological Investiga- . tions of the Brickell Point Site (8DA12), Including the Miami Circle. The Florida Anthropologist 53:260-293. Carrasco-Niifiez and W.I. Rose