Arkansas Historic Preservation Program
Blue Spring Shelter
Blue Spring Shelter

BLUE SPRING SHELTER, RESTRICTED, CARROLL COUNTY

SUMMARY

The Blue Spring Shelter is being nominated to the National Register of Historic Places with statewide significance under Criterion D in that it is likely to yield information important in prehistory. There are four major occupations periods evident in the Blue Spring Shelter. These are the Early Archaic, Middle Archaic, Late Archaic and Mississippian. All these periods are imperfectly known in Northwest Arkansas (see Sabo, et al 1990 and Raab, et al. 1982). Blue Spring Shelter contains well-preserved deposits that can answer important questions about Early and Middle Archaic subsistence, the domestication of native plants and the development of horticulture in the Ozarks during Late Archaic times, and the stability of Mississippian settlements in the narrow river valleys of northwest Arkansas.

ELABORATION

The Archaic period in Arkansas history extends from 7,500 to 500 BC. During this period the climate adjusted to the retreat of glacial ice in the north, became actually warmer and dryer than today from 6,000 to 2,000 BC, before the essentially modern climate and vegetation was finally established around 2,000 BC. For many years, archeologists defined Archaic sites and peoples by what they did not have compared to the later time periods. It was thought that Archaic peoples did not make mounds or pottery. It was also thought that Archaic peoples did not grow their food. Today, we know all of these major innovations were developed by Archaic peoples. However, we do not know the exact timing of these events, and how climatic change, plant domestication, population increases, and the changing locations and nature of their settlements affected these developments. The information in Blue Spring Shelter will not answer all of these questions, but it will provide important information about them and refine and focus future research questions.

It is assumed that Early Archaic (7500 – 6000 BC) people in Arkansas lived in small bands of 25 to 30 people, and moved their camps frequently taking advantage of the changing seasonal abundance of wild plants and animals, but little is known about how people lived in this time period (Sabo et al. 1990). Early Archaic components have been excavated at the other shelters in Arkansas, notably at the Tom’s Brook Shelter (Bartlett 1963), Albertson Shelter (Dickson 1991) and Breckenridge Shelter (Wood 1962, Thomas 1969). The recovered artifacts, dart points and knives, seem to reflect small hunting camps. The same is true for the few open sites were Early Archaic artifacts have been found (Lafferty et al. 1988; Mainfort 2000). However little is known about exactly what they were hunting or their seasonal movements. The well-preserved animal bone and dry deposits at the Blue Spring Shelter will provide important information on basic subsistence of Early Archaic peoples in Northwest Arkansas and the nature of their social groups.

It has been suggested (Sabo et al 1982; Sabo et al 1990) that there was change in the settlement systems of Archaic peoples during the Middle Archaic period (6,000 – 2000 BC) from small camps to larger more sedentary settlements in the river valleys. This would correspond to the warmer and dryer climate characteristic of the Mid-Holocene. If this were true, Blue Spring Shelter would be an ideal location for a base camp as it provides shelter adjacent to the White River flood plain. Little is known about subsistence and seasonal movements of Middle Archaic peoples or the nature of their villages. It has been suggested that during this period, due to the extended warm and dry period, deer populations in Arkansas would have been reduced forcing people to rely more on small game and river resources such as fish, river mussels and turtles (Sabo et al 1990; Raab et al. 1982). Blue Spring Shelter can provide important information to answer these questions because of the excellent preservation of animal bone and the presence of food processing features.

One of the most significant events in the Native American history of Arkansas and the Mississippi Valley is the domestication of plants native to the region. These include squash (Curcurbita spp); starchy seeds like goosefoot (Chenopodium berlandieri), little barley (Hordeum pusilum), and maygrass (Phalaris caroliniana); and oil rich seeds such as marshelder (Iva annua) and sunflowers (Helianthus annus). Goosefoot, marshelder, sunflower, and squash were domesticated prior to 1,000 BC during the middle of the Late Archaic time period. The others were domesticated sometime between 1,000 BC and AD 300 (Smith 1992).

Bruce Smith (1992) argues these plants were initially domesticated in Archaic settlements located on the floodplains of large rivers. His logic is that all of these plants invaded disturbed soils (today they are considered weeds) which would be common in a river valley scoured by floods and near villages where people were creating soil disturbances by building houses, digging pits, and cutting trees. The soils around these villages are especially fertile because of trash disposal. Smith argues there are three stages in the domestication of these plants. Initially people would harvest the wild seeds close to the village. Next they would actively encourage and care for the still wild plants by removing competitive plants (weeding). Finally, the seeds would be intentionally planted. Domestic seeds can be identified in archeological sites because the seed morphology changes. Domestic seeds are larger and the protective coating (called the testa) thinner than their wild counterparts.

These native domesticated plants were commonly found in the Ozark bluff shelters excavated in the 1930’s by the University of Arkansas Museum (Gilmore 1930; Fritz 1984, 1986, 1997). Unfortunately the excavators who recovered these plants were searching primarily for basketry, matting, arrow shafts and other perishable artifacts for museum exhibits. The plant remains were recovered almost accidentally as they were found in small bags and other containers. Also, in most instances, these early excavations did not go deep enough to reach Archaic deposits. So while we know Arkansas Indians were growing these native domesticated plants, we do not know when or how they acquired them. Based on information from surrounding states they should have acquired them or were active participants in their domestication during Middle and Late Archaic time periods. Blue Spring Shelter not only contains extremely well preserved Middle and Late Archaic deposits but it is located in floodplain where early plant domestication most likely occurred. The Archaic deposits in the Blue Spring Shelter can provide critical information about plant domestication and/or the adoption of domesticated plants in the Ozarks.

While seeds and other plant remains have not been recovered from Blue Spring Shelter it is almost certain they exist in the deposits in datable contexts. When excavations were conducted at the Blue Spring Shelter in 1970 and 1971, archeologists were in the initial stages of developing water flotation techniques for the recovery of small seeds from archeological sites and Chenhall did not attempt to recover small seeds. The only way he would have recovered seeds was if they were preserved in small bags or other containers. Today archeologists have water flotation and separation equipment that can recover the extremely small seeds of even tobacco and chenopodium. Use of this water flotation equipment is how archeologists know these same native plants were grown at Toltec Mounds and other open sites where the preservation environment is less favorable than Blue Spring Shelter.

Based on the numbers and size of recorded sites it appears the population greatly increased during Late Archaic times (2000 – 500 BC) in the Ozarks (Sabo et al. 1990). A more favorable climate may be one explanation for the population increase, but it also may be that horticulture provided some stability to a subsistence economy based on the hunting and gathering of wild foods whose abundance fluctuated year to year. The relationship between plant domestication, increased population, and sedentary habits is a fundamental research topic in modern archeology, and it is fundamental to our understanding of Ozark history. The archeological deposits at the Blue Spring Shelter can provide critical information about the nature of Archaic life and settlement in at least one river valley in northwest Arkansas.

In contrast to northeast Arkansas, Mississippian (AD 900 - 1700) settlements in northwest Arkansas tend to be small and dispersed in the river valleys, but a few larger sites with multiple mounds are located both in the Illinois and White river drainages (Sabo et al. 1990). The people grew corn, squash, possibly beans, and maybe still grew some of the native domesticates characteristic of earlier time periods. Produce from their farms was supplemented with wild plants foods, especially acorns and other nuts (Hilliard 1982). Deer, turkeys and fish provided basic protein. Most rock shelters in the Ozarks contain sizeable Mississippian components. Archeologists believe rock shelters were the locations of temporary hunting and gathering camps used by Mississippian peoples who were hunting deer and other animals and gathering various varieties of nuts and berries. However, there are still critical questions about the stability of Mississippian villages in the narrow river valleys of the Ozarks. It is suggested that unlike the Mississippi River Valley where Mississippian peoples were dependent on maize agriculture, the people who lived in the Ozarks may have been more dependent on wild foods, especially nuts foods. If this is true rock shelters would be important locations for hunting and gathering and the archeological record in the shelters may reflect larger numbers of people using and living in the shelters than just simple hunting camps. The Mississippian deposits in the Blue Spring Shelter contain hearths, arrow points, food remains and other evidence that will help answer this question.

The deposits at Blue Spring Shelter do contain evidence of use during the mid-19th Century presumably by people who operated the mill near the site. A small building may have existed in the shelter as evidenced by square nails found in the 1970 excavations. In addition, bottle glass, pottery and various scraps of metal were also found. Due to the disturbances in the upper levels of the shelter this component is not believed to have archeological integrity and does not contribute to the National Register significance of the shelter.

SIGNIFICANCE

The Blue Spring Shelter is being nominated to the National Register of Historic Places with statewide significance under Criterion D in that it is likely to yield information important in prehistory. There are four major occupations periods evident in the Blue Spring Shelter. These are the Early Archaic, Middle Archaic, Late Archaic and Mississippian. All these periods are imperfectly known in Northwest Arkansas (see Sabo, et al 1990 and Raab, et al. 1982). Blue Spring Shelter contains well-preserved deposits that can answer important questions about Early and Middle Archaic subsistence, the domestication of native plants and the development of horticulture in the Ozarks during Late Archaic times, and the stability of Mississippian settlements in the narrow river valleys of northwest Arkansas.

BIBLIOGRAPHY

Bartlett, Charles S., Jr. 1963 The Tom’s Brook Site – 3J01 – A Preliminary Report, in Arkansas Archeology, 1962, edited by Charles R. McGimsey III, pp. 15-65. Arkansas Archeological Society, Fayetteville.

Chenhall, Robert G. 1971 Spring Shelter: A Preliminary Report. Field Notes of the Arkansas Archeological Society, No. 74, February, pp 9-12.

Delcourt, H. R., and P. A. Delcourt 1994 Postglacial rise and decline of Ostrya virginiana (Mill.) K. Koch and Carpinus caroliniana Walt. in eastern North America: predictable responses of forest species to cyclic changes in seasonality of climates. Journal of Biogeography 21:137-150.

Dickson, Don R. 1991 The Albertson Site. Arkansas Archeological Survey Research Series No. 41. Fayetteville.

Fisher-Carroll, Rita 2001 Environmental Dynamics of Drought and its Impact on Sixteenth Century Indigenous Populations in the Central Mississippi Valley. Ph.D. dissertation, Environmental Dynamics Program, University of Arkansas, Fayetteville.

Fritz, Gayle J. 1984 Identification of cultigen Amaranth and Chenopod from Rockshelter Sites in Northwest Arkansas. American Antiquity 49:558-572.

1986 Prehistoric Ozark Agriculture – the University of Arkansas Collections. Ph.D. dissertation, University of North Carolina at Chapel Hill.

1997 A Three Thousand Year Old Cache of Crop Seeds from Marble Bluff, Arkansas. In People, Plants, and Landscapes: Studies in Paleoethnobotony, edited by K.J. Gremillion, pp. 42-62. University of Alabama Press, Tuscaloosa.

Gilmore, Melvin R. 1930 Vegetal Remains of the Ozark Bluff Dweller Culture. Papers of the Michigan Academy of Science, Arts, and Letters. 14:83-102.

Hilliard, Jerry E. 1986 Selection and Use of Acorn Species by Late Prehistoric Ozark Inhabitants. In Contributions to Ozark Prehistory, edited by George Sabo III, pp. 81-85. Arkansas Archeological Survey Research Series No. 27, Fayetteville.

Lafferty, Robert H. III, Larry G. Santeford, Margaret J. Guccione, Michael C. Sierzchula, Neal H. Lopinot, Kathryn King, Kathleen M. Hess, and Jody O. Holmes 1988 The Mitchell Site: 3WA58 Archeological Investigations at a Prehistoric Open-Field Site in Washington County, Arkansas. Mid- Continental Research Associates, Report 87-4. Lowell, Arkansas.

Lopinot, Neal H., Jack H. Ray, and Michael D. Conner (editors) 1998 The 1997 Excavations at the Big Eddy Site (23CE426) in Southwest Missouri. Special Publication No. 2, Center for Archaeological Research, Southwest Missouri State University. Springfield, MO.

Mainfort, Robert C. 2000 Data Recovery at the Skaggs Site, Madison County, Arkansas. Arkansas Archeological Survey Research Report No. 28. Fayetteville, Arkansas.

Mainfort, Robert C., Jr., Michael M. Evans, and Jared S. Pebworth 2002 A Phase I Cultural Resources Survey of the Proposed City of Fayetteville Wastewater Treatment Facility Site, Associated Water Lines, and Lift Stations, and the Proposed Extension of Broyles Road, Washington County, Arkansas. AAS Project Numbers 02-14 and 02-16 Draft Report. Submitted to McGoodwin, Williams and Yates, Inc., Fayetteville, Arkansas

Raab, L. Mark, Gayle J. Fritz, Daniel Wolfman, Robert H. Ray, and George Sabo, III 1982 The Arkansas Ozarks. In, A State Plan for the Conservation of Archeological Resources in Arkansas, edited by Hester A. Davis, pp. NW1- NW24. Arkansas Archeological Survey Research Series No. 21, Fayetteville, Arkansas.

Sabo, George, III, David B. Waddell, and John H. House 1982 A Cultural Resource Overview of the Ozark-St. Francis National Forests, Arkansas. Arkansas Archeological Survey, submitted to the USDA Forest Service.

Sabo, George, III, Ann M. Early, Jerome C. Rose, Barbara A. Burnett, Louis Vogele, Jr., and James P. Harcourt 1990 Human Adaptation in the Ozark and Ouachita Mountains. Arkansas Archeological Survey Research Series No. 31.

Smith, Bruce D. 1992 Prehistoric Plant Husbandry in Eastern North America. In The Origins of Agriculture: An International Perspective, edited by C. Wesley Cowan and Patty Jo Watson, pp 101-119, Washington DC, Smithsonian Institution Press.

Thomas, Ronald A. 1969 Breckenridge: A Stratified Shelter in Northwest Arkansas. MA Thesis, University of Arkansas.

Wood, W. Raymond 1963 Breckenridge Shelter – 3CR2: An Archeological Chronicle in the Beaver Reservoir Area. In Arkansas Archeology, 1962, edited by Charles R. McGimsey III, pp. 67-96, Arkansas Archeological Society, Fayetteville.