2006 Pennsylvania Archaeologist 76(1):1-19
A Preliminary Report on the Ebbert Spring Site
Ronald D. Powell

In July 2003 members of SPA Chapter No. 27 began excavating the Ebbert Spring site (36FR367) located in Franklin County, Pennsylvania. Artifacts and settlement patterns indicate that the Ebbert Spring site served as a prehistoric camp from Paleoindian times until the Contact period. The site continued to be of importance in the early historic period, with the original springhouse probably serving at times as a settlers' fort. Recovered prehistoric materials indicate a possible trade network along the Virginia Path from the Potomac and Susquehanna rivers. Future excavations should continue to expand our understanding of the site's interesting history.
 

Excavations in 1988 at Horseshoe Rockshelter (36CH488) demonstrated that Native Americans utilized the shelter from Early Archaic times through the Contact period. Evidence suggests that the site was primarily used as a short term hunting camp during prehistoric times. There also is evidence that the site functioned as a quartz procurement/processing camp, most likely during the Late Woodland period. The shelter continued to experience periodic use by Euroamericans during the 19th and 20th centuries. Boy Scout camping activities at the rockshelter in recent times disturbed much of the prehistoric cultural deposits. Nevertheless, the excavation produced significant data regarding prehistoric activities at the site.
 

Prehistoric settlement patterns and lithic raw material usage in the Upper Juniata er basin of south-central Pennsylvania are examined and discussed. Results of the study indicate that substantial population shifts occurred in the study area over the course of the Holocene, including two peaks in site counts during the Late Archaic and  Late Woodland periods, respectively. These site count changes likely correlate with actual  population fluctuations in the Upper Juniata River basin, rather than with changes in site  visibility over time. Lithic raw material data indicate that a variety of local cherts were favored during the  Holocene. Rhyolite was the main type of non-local toolstone used in the study area, with a  peak in use during the Late Archaic period. According to regional models, during the  Late Woodland period the use of villages and horticulture increased in central  Pennsylvania; however, little evidence of such trends is present in the study area.  Instead, the region was likely a cultural buffer zone used by multiple ethnic groups for  hunting and gathering.
 

A review by the author of radiocarbon dates recently used by George (2004a, 2004b, and 2004c) to ad ance his concept of the Drew "tradition" revealed a number of discrepancies and contextual problems. In particular, dates that were used to extend the temporal span of the Drew phase into the Middle and Late Monongahela periods, which in turn, suggested a reconceptualization of the Monongahela culture, proved most problematic. These issues are discussed, as well as the possible implications regarding our understanding of Drew cultural development.
 

The Caneadea Mound was unscientifically excavated by artifact collectors in 196 3 . Although not directly involved in the excavation, the late Dr. Marian White documented much of the material contents of the mound, resulting in one of the most extensive mound data sets available in the Northeast. The artifacts and a radiocarbon date indicate that the mound was constructed in the Middle Woodland period. While the artifacts suggest a relationship to mounds in the Ohio River Valley, they also demonstrate regional differences. Evidence from this and other mounds in the Northeast suggests a broad-reaching, shared burial tradition, in place since the Late Archaic period. This shared tradition may have formed the foundation for various cultural developments, including the Ohio Hopewell. Elements of the Ohio Hopewell fluorescence are evident at the Caneadea and other northeastern mounds, but direct Hopewell influence appears to have been minimal. Data from northeastern mounds indicate that Hopewell may not be appropriate as a universal label for Middle Woodland mound building cultures.