Vermont Archaeology Comes of Age: A Perspective on Vermont ... · Introduction Throughout prehistory people had to adapt to the world around them. This is not to say that the environment

Vermont Archaeology Comes of Age:A Perspective on Vermont's Prehistoric Past

By Peter A. Thomas

Abstract

An estimated 200 archaeological surveys, site evalua-tions, and more extended excavations have been conduct-ed in Vermont during the past 15 years. As a result,considerable information has been gathered about manyaspects of Vermont's prehistory. This paper has twogoals: to synthesize the information gained from thesestudies and to integrate this information into a context ofwhat is currently known about Vermont's prehistoric past.

The text is divided into eight sections. The first providesan overview of changing plant and animal communitiesand the various climatic periods which affected peoples'lives during the past 12,000 years. Remaining sectionssummarize what we can currently infer about life duringthe Paleoindian, Archaic and Woodland periods. Eachsection addresses four specific topics: regional context (toprovide an understanding of how Vermont sites relateto a wider area), technology, settlement and subsistence,and ideology. This text is an updated, but much abbre-viated version of Our Prehistoric Cultural Heritage, theprehistoric context of the Vermont Historic PreservationPlan which I prepared for the Vermont Division forHistoric Preservation in 1991.

PALEOENVIRONMENTS-AN OVERVIEW

Introduction

Throughout prehistory people had to adapt to the worldaround them. This is not to say that the environmentforced people to make specific decisions, but the envi-ronment did place limits on the number of choices whichcould be made. Many of the inferences which archaeolo-gists make about how people lived in the past are basedheavily on reconstructions of regional and local environ-ments. It is unclear when small communities of peoplefirst utilized the natural resources of Vermont or exploredthe more attractive areas. Based on current evidence, thefirst Native Americans probably arrived between 11,300and 10,500 B.P.

Probably the most dramatic changes in the landscapeclosely followed glacial retreat. Between 12,500 and10,200 B.P., the land rose as the pressure produced by themassive weight of the glacier was released; the surfacetopography was greatly modified as streams and riversformed and cut deep channels; soils began to develop; var-ious plant communities became established; a variety ofanimal, bird, and fish species spread into favorable habi-tats; an arctic climate became less severe; and, in theChamplain Valley, the salt water Champlain Sea drainedand was replaced by Lake Champlain. At its inception,Lake Champlain was 15-20 feet lower in elevation thanit is today, and, at its northern end, had a substantially dif-ferent configuration (Fillon 1970:26-30). Since thattime, resident groups of people have adapted to a slow-ly but constantly changing environment. The landscapecontinued to be reshaped; soils continued their develop-ment; various animals and plants spread into new terri-tories or even disappeared as broad climatic changesoccurred; the level of Lake Champlain gradually rose,and lake waters flooded former shore zones (Beblowski1981:487).

One of the goals of future archaeological research inVermont will be to greatly expand the types of environ-mental information which are currently available, but sub-stantial gains in understanding various aspects of environ-mental change have been made in the past fifteen years.The following section on Vermont'schanging forestenvironment, climate and an mal communities provides ageneral overview which incorporates information fromrecent studies (Figure 1).

Changing Patterns

As an outgrowth of pollen and other studies from sitesthroughout the Northeast and broader continental UnitedStates, researchers have attempted to reconstruct the grosschanges in general climatic patterns during the past 12,000years. The underlying assumption has been that there isa positive correlation between climatic trends and changes

38

The Journal of Vermont Archaeology

Figure 1. Generalized environmental change since 12,500 B. P.

YEARSB.P.

REGIONALVEGETATION

TYPE

CLIMATI.CTRENDS

ALLUVIALEPISODES

ALONG RIVERS

FAUNA MARINE/LAKEFAUNA

CULTURALPERIODS

1000

2000

3000

4000

5000

6000

7000

8000

9000

10,000

11,000

12,000

13,000

Oak-ChestnutForest

--------------oak-Hickory

Forest(Oak, Pine,

Beech Maximum)

--------------Oak-Hemlock

Forest

(HemlockBeech

Maximum)

--------------Pine DominatedForest (Oak,Hemlock, Elm,Ash, poplar)--------------Spruce -

FirWoodland

Tundra

BarrenGround

Series ofWarmer -Cooler,Wetter -Drier

Episodes

WarmestDry

WarmerMoist

TemperatureExtremesLessen

Dry, WarmSummers

Dry, ColdWinters

SeasonalExtremes

Subarctic

FrigidArcticClimate

VERMONT

CyclicalEpisodes

ofRapid &Slower

Alluviation

AlluviationSlow

InfrequentFlooding

MajorFlooding

RapidA11uviation

AlluviationSlow

--------------Modern Valley

BottomsEstablished

RapidAlluviation

MassiveDown Cutting

of RiverChannels

COVERED

ModernSpecies

DeerBear

MooseBeaverMartinLynx

MuskoxCaribouMammothMastodon------------

BY

ModernSpecies

Catfish

WhitefishLake Trout

Lake ChamplainFormed

Salmon/SmeltWhalesSeals

Champlain SeaFormed

GLACIER

HistoricPeriod

LateWoodland

MiddleWoodland

EarlyWoodland

LateArchaic

MiddleArchaic

EarlyArchaic

Paleo indian

39

Vermont Archaeology Comes of Age

in regional vegetation patterns recorded in pollen cores,regional lake levels, cyclical changes in the earth's axis,atmospheric insolation, and other factors.

Vermont was dominated by glacial ice prior to 13,500B.P. By roughly 12,600 B.P., the glacier had moved northof the St. Lawrence lowland, leaving behind a landscapeof frozen ground for most of the year. A frigid arcticclimate existed, with the prevailing winds coming out ofthe north and northeast, and both the summer and winterArctic Frontal Zone of the continental air mass lying to thesouth of the Champlain Valley (Ogden 1977:19-20;Shafer and Hartshorn 1965:124). A tundra environmentof mosses, lichens and stunted shrubs characterized thelandscape. Between roughly 11,300 and 10,000 B.P.,a subarctic climate dominated the region. Subsequently,a shift to a more temperate climate seems to have begun,but weather patterns were still greatly affected by the icesheet to the north (COHMAP Members 1988; Ogden1967; Webb and Clark 1977).

In Vermont, and particularly in northern Vermont and athigher elevations, this was generally a period of transitionfrom a predominantly tundra environment of sparse vege-tation to an open park-like woodland as spruce, fir andbirch began to take hold. This is also the period when peo-ple first began to move into Vermont. A spruce-firwoodland was well established in Vermont's lowlands bylO,200 B.P. and had begun to expand into the mountainsby at least 9500 B.P. Larch and alder were present in lowpercentages in the wet lowlands, and beech, oak, ash andmaple began to appear on the better drained bottomlandareas and low hills in the Champlain and ConnecticutValleys.

As the glacier receded after 12,500 B.P., wildlife adaptedto a cold environment migrated to Vermont. Beginningwith the establishment of a tundra environment and subse-quently by the spread of open spruce parklands into thelowlands of Vermont, the region supported a population ofwoolly mammoths, mastodons, moose, elk, and herds ofcaribou. As recently as 1989, a mammoth tooth wasrecovered from a stream bed in Derby, Vermont. Muskoxand smaller arctic animals, such as ptarmigan, arcticshrews and lemmings, may have also been present (Parren1988). As stands of spruce, fir, aspen, alder, and birchgrew up, moose, beaver, lynx, porcupine, snowshoe rabbit,spruce grouse, mice and voles probably migrated from thesouth. Beaver-gnawed wood has been recovered fromearly bog deposits near Stowe, and the remains of hare,eastern chipmunk and American marten have been recov-

ered from deposits in Canada which are older than 10,200B.P. (Harrington 1977).

The Champlain Sea may have been of particular impor-tance for people moving into northwestern Vermont. Priorto 10,200 B.P. when it finally drained, the ChamplainSea may have provided an abundance of food resources notavailable at a later time. The distribution of classicPaleoindian sites (ca. 11,000-10,000 B.P.) in areas closeto the margins of the sea suggests that marine resourcesmay once have been exploited (Loring 1981). The recordof marine fossils from Champlain Sea deposits indicatesthe presence of large marine mammals between roughly11,600 and 10,200 B.P. White, common finback, hump-back and bowhead whales, possibly harbor porpoise, aswell as harp, ringed, and hooded seals have been found atdifferent locations, including the Champlain Valley. Aconvincing case has been made for the arrival of Atlanticsalmon and two varieties of smelt in the Vermont arm ofthe Champlain Sea during its final phase. By about 10,300B.P., with the establishment of near-fresh water condi-tions in the Champlain Valley, it appears likely that laketrout, whitefish and brook trout were also becoming estab-lished (Brooks and Deevey 1963:144-152).

It should be pointed out that archaeologists do not cur-rently know whether Paleoindian populations ever foundways to utilize marine resources. If marine hunting andfishing strategies were adopted in New England, howev-er, the only surviving evidence may be found in the archae-ological record of the Champlain Lowland. On theAtlantic coast, contemporaneous sites which may havebeen the bases of such operations are now submergedsome 30-90 miles off shore.

By approximately 9000 B.P., a sudden sharp increase inwhite pine, hemlock, oak, poplar, elm, ash, sweet galeand ferns in pollen profiles marks the establishment ofa true forest environment. The establishment of apine-dominated forest provides definite confirmation ofa warming trend. Precipitation rates seem to have beenfairly low, a characteristic which appears to be reflectedby the presence of very thin alluvial beds which character-ize some floodplains dating to this period (Thomas andDillon 1983). A trend of dry, warm summers and dry, coldwinters seems to be characteristic throughout the north-ern hemisphere at this time (Dincauze 1989). Seasonaltemperature extremes may have been characteristic.Current estimates are that average solar radiation was 8° Chigher in July and 8° C lower in January when comparedto today's extremes. After 9000 B.P., these seasonal radi-

40

ation extremes decreased towards more modem values(COHMAP Members 1988).

With the establishment of a closed woodland around9000 B.P., it has been presumed that with the lower car-rying capacity of a coniferous woodland most of the largegame animals disappeared and that the forest environ-ment of Vermont and northern New England in generalwas impoverished. This assumption was once used toexplain why few archaeological sites dating to this periodhad been identified (Fitting 1968; Ritchie 1965), butthis position no longer seems tenable. Rather, it is likelythat boreal animals, such as moose, beaver, lynx, porcu-pine, snowshoe rabbit and spruce grouse, were present ina number of favorable habitats throughout the state asearly as 9500 B.P. There is also increasing evidence thatpeople continued to live throughout northern NewEngland at this time as well (Robinson, Petersen andRobinson 1992).

In sediments dated to 8000 B.P., the pollen complex indi-cates the presence of a temperate forest characterized by ahigh percentage of oak, beech, birch, sugar maple, elm andash. Hemlock and white pine were the dominant conifers.Within this mixed deciduous-coniferous forest in the low-lands of eastern and western Vermont, deer, bear, wolf,raccoon, otter, fox, grey squirrel, wild turkey and passen-ger pigeon occupied favorable habitats. Moose, elk andperhaps even small herds of woodland caribou were prob-ably more common in the spruce-fir-northern hardwoodforests at higher elevations in the Piedmont and GreenMountains. Such upland and lowland game animalsformed the foundation of a hunting-fishing-gatheringeconomy which supported Native American communitiesin Vermont from the end of the Early Archaic period tothe time of European exploration.

With the disappearance of the Champlain Sea and theestablishment of a freshwater environment in LakeChamplain after 10,200 B.P., the types, densities and diver-sity of aquatic resources also changed. Species such asbrook trout, perch, pike, chub and mullett expanded intothe rivers, smaller streams and ponds. At least seasonally,fish may have been an important source of food for peopleliving in the Champlain Valley, as well as in theConnecticut and Hudson River watersheds in easternand southwestern Vermont, beginning some 8,000-10,000 years ago. A radiocarbon date of 8700 B.P. inassociation with fish bone at the WMECO site in the mid-dle Connecticut Valley and the recovery of a pectoralspine from a catfish at the John's Bridge site on the


Missisquoi River in northwestern Vermont dated to 8100B.P. support this inference (Curran and Thomas 1978;Thomas and Robinson 1980).

Based on changes in the percentages of various plantspecies, pollen cores reflect various periods of regionalchange in the forest environment. These periods havebeen termed oak-hemlock (8000-5000 B.P.), oak-hicko-ry (5000-3100 B.P.) and oak-chestnut (3100 B.P. to thepresent). In culture history terms, these intervals corre-spond to periods which run from roughly the end of theEarly Archaic to the early Late Archaic perod, for theremainder of the Late Archaic period, and from the EarlyWoodland through the Late Woodland period. Whilepollen data are useful for understanding general trendsin the environmental development of the region, thequantum leap between establishing regional profiles anddefining the characteristics of local plant populationsthrough time has not been made. These schemes do notreflect the dynaics within forests where any or all plantpopulations may grow, remain stable, or decline withtime. Given the generally rugged topography ofVermont, the substantial changes in elevation from eastto west, and the evolution of plant communities through-out the past 8,000 years, the term "patchy" most aptlydescribe the forested landscape of Vermont. This patch-iness offered great opportunities for human exploitation.How people operated within various environmental set-tings will undoubtedly be reflected in Vermont's richarchaeological record.

By approximately 7500 B.P., or the beginning of theMiddle Archaic period, the severe climatic conditions ofthe previous period began to ameliorate. Although theenvironmental data need to be considerably refined forthe New England region, various peripheral sources sug-gest that the period from 7500 to roughly 5300 B.P. wasone of fairly warm and moist conditions. Precipitation mayhave been as much as 25-30% higher than at present.Major flooding was common along a number of river sys-tems. These trends are suggested by the occurrence of ahemlock and beech maxima in Vermont pollen profiles, aswell as by evidence of rapid sedimentation and channelmigration along stretches of the Missisquoi River between6500 and 5400 B.P. and by the presence of beech, cedar,maple and hemlock logs entombed in the Missisquoi Riverfloodplain in Highgate which date to the same period(Brakenridge, et a1. 1988; COHMAP Members 1988;Thomas and Dillon 1983; Thompson and Bettis 1982). Thesubsequent period, dating from approximately 5300 to2900 B.P., is marked by a climatic shift to drier condi-

41


tions. Recent studies suggest that average summer tem-peratures were from 20 to 40 C higher than at present,while temperature extremes during the winter relaxed(COHMAP Members 1988; Dincauze 1989). In the upperMidwest, Thompson and Bettis (1982) found this to begenerally a period when major rivers entrenched andflooding was infrequent; prairies reached their maximumextent. The apparent lack of any substantial alluvialdeposits dating to this period within the floodplains studiedalong the Missisquoi River seems to mirror this episode oflimited flooding and channel entrenchment (Brakenridgeet al. 1988; Thomas and Dillon 1983). A significantdecrease in precipitation at this time is further suggestedby a substantial drop in the water table of Shelburne Pond,located in the Champlain Lowland (Carr, Worley andDavis 1977). Cyclical fluctuations of climatic trendsafter 2800 B.P. seem to be better defined. The pollenrecord indicates that there were a series of warmer-to-cooler and wetter-to-drier episodes within the past2,800 years (Bernabo 1981; Swain 1978; Wendlund andBryson 1974). Voight and O'Brien (1982) summarize thesequence of climatic shifts as follows.

Between 2800 and 1680 B.P., much of the midcontinentwitnessed a general relief from hotter and drier condi-tions. As annual temperatures dropped and precipitationincreased in Vermont, a resurgence of spruce and firoccurred at higher elevations; pine began to increase in fre-quency in the lowlands. Between 1680 and 1200 B.P.,warmer temperatures blanketed the central United Statesand oak-dominated forests maintained relatively stabledistributions. Around 1200 B.P., precipitation levelsincreased throughout the Northeast. Temperatures werestill warm.

With the onset of the next climatic episode at approxi-mately 900 B.P., at the beginning of the Late Woodlandperiod, increased penetration by pacific air masses dur-ing the summer months resulted in droughty conditionsand higher temperatures. The period between 950 and750 B.P. may have been the warmest time during the past2,000 years. Changes in temperature and moistureregimes resulted in lowered water tables and a decrease instream flow and frequency and duration of flooding. Thepeak of this episode occurred around 600 B.P. A shift tocooler and moister conditions began about four hundredyears ago and continued until the late nineteenth century.

Substantial changes in regional climatic trends haveoccurred throughout the past 12,000 years. More subtlecontrasts in temperature and precipitation undoubtedly

existed among Vermont's physiographic regions, as well.Both broad and small-scale climatic patterns are likely tohave played important roles in the lives of Vermont'sprehistoric inhabitants. Much research remains to be com-pleted before more can be said about the first occur-rences of various animals, their distributions and how therange and size of various populations fluctuated throughtime. Given the pronounced climatic swings and changesin forest types noted previously, substantial changes inmammal, reptile, amphibian, bird and fish populationsmust have also occurred since the forests first closed in onthe Vermont landscape. We must recognized and under-stand such changes before we can gain a fuller appreciationof the lifeways of people who lived in Vermont during thepast 10,500 years.

Within the past 200 years, climatic changes and humanactions have substantially altered the landscape NativeAmericans once knew. Since the late nineteenth century,the climate has ameliorated and we are now faced withglobal warming and stunted forest growth due to industri-al pollution. A number of animal species have beenexterminated from Vermont and massive areas of forestland were cut over during the last century and projectsundertaken throughout Vermont (Doherty, Kochan andThomas 1992; Thomas 1985a, 1985b, 1989a, 1989b, 1991;Thomas and Bourassa 1985; Thomas, Clement andDoherty 1992)

Understanding the relationship between the natural envi-ronment and human cultures has been and will continue tobe one of the fundamental goals of archeological research.

Paleoenvironments: References

Beblowski, P.1978 Geology of the Winooski Site. In The WoodlandPeriod in the Northeast, by J. B. Petersen and M. Power,pp. 456-493. Report for Interagency ArchaeologicalServices, Philadelphia. Department of Anthropology,University of Vermont, Burlington.

Bernabo, J. C.1981 Quantitative Estimates of Temperature ChangesOver the Last 2,700 Years in Michigan Based on PollenData. Quaternary Research, Vol. 15:143-159.

Brakenridge, G. R., P. A. Thomas, L. E. Conkey, and 1.Schiferle1988 Fluvial Sedimentation in Response to PostglacialUplift and Environmental Change, Missisquoi River,

42

Vermont. Quaternary Research, Vol. 30: 190-203.

Brooks, 1. L. and E. S. Deevey1963 New England. In Limnology in North America, edit-ed by D. G. Frey, pp. 117-162. Madison: University ofWisconsin Press.

Carr, P. G., I. A. Worley and M. W. Davis1977 Post-Lake Vermont History of a Pond and Wetlandin the Champlain Basin. In Proceedings of the 4thAnnual Lake Champlain Basin EnvironmentalConference, pp. 71-112. Institute for Man andEnvironment, SUNY, Plattsburg.

COHMAP Members1988 Climatic Changes of the Last 18,000 Years:Observations and Model Simulations. Science, Vol.241: 1043-1052.

Curran, M. L. and P. A. Thomas1979 Phase III Data Recovery: Wastewater TreatmentSystem in the Riverside Archaeological District ofGill, Massachusetts. Department of Anthropology, Report19. University of Vermont, Burlington.

Dincauze, D. F.1989 Geoarchaeology in New England: An EarlyHolocene Heat Spell? The Review of Archaeology, Vol.10(2):1-4.

Doherty, P., G. Kochan and P. A. Thomas1992 Archaeological Site Identification Survey forDuxbury 2809, Duxbury, Vermont. ConsultingArchaeology Program, Department of Anthropology,Report 118. University of Vermont, Burlington.

Fillon, R. H.1970 The Sedimentation and Recent Geologic History ofthe Missisquoi Delta. MS thesis, Department of Geology,University of Vermont, Burlington.

Fitting, J. E.1968 Environmental Potential and the PostglacialReadaptation in Eastern and the Postglacial Readaptationin Eastern North America. American Antiquity, Vol.33(4):441-445.

Harrington, C. R.1977 Marine Mammals in the Champlain Sea and theGreat Lakes. In Amerinds and Their Paleoenvironmentsin Northeastern North America. edited by W. S. Newman


and B. Salwen, pp. 508-537. Annals of the New YorkAcademy of Sciences 288.

Loring, Stephen1980 Paleo-Indian Hunters and the Champlain Sea: APresumed Association. Man in the Northeast, Vol. 19:15-42.

Ogden, J. Gordon, ill1967 Radiocarbon and Pollen Evidence for a SuddenChange in Climate for the Great Lakes RegionApproximately 10,000 Years Ago. In QuaternaryPaleoecology, edited by E. J. Cushing and H. E. Wright,Jr., pp. 117-127. New Haven, Yale University Press.

Parren, S.1988 Wildlife in Vermont's Changing Landscape: A13,000 Year Perspective. Habitat Highlights, Vol. 8(4):1-4.

Ritchie, W. A.1965 The Archaeology of New York State. Garden City:Natural History Press.

Robinson, B. S., J. B. Petersen and A. K. Robinson1992 Early Holocene Occupation in Northern NewEngland. Occasional Publications in Maine Archaeology,Number Nine. Augusta, Maine.

Shafer, J. P. and J. H. Hartshorn1965 The Quaternary of New England. In The Quaternaryof the United States, edited by H. E. Wright, Jr. andD. G. Frey, pp. 113-128. Princeton: Princeton UniversityPress.

Swain, A. M.1978 Environmental Changes During the Past 2,000Years in North-central Wisconsin: Analysis of Pollen,Charcoal, and Seeds from Varied Lake Sediments.Quaternary Research, Vol. 10:55-68.

Thomas, P. A.1985a Archaeological and GeomorphologicalEvaluation, Burlington M5000(3) Northern Connector.Material Supply/Disposal Area, Howe Farm Floodplain.Department of Anthropology, Report 54. University ofVermont, Burlington.

1985b St.Johnsbury FEGC DP EF 041-1(10), CaledoniaCounty, Vermont: Archaeological Reconnaissance Survey.Department of Anthropology, Report 57. University of

43


Vermont, Burlington.

1989a Preliminary Archaeological Survey,Environmental Conservation and AgriculturalLaboratory, Waterbury, Vermont. Department ofAnthropology, Report 100. University of Vermont,Burlington.

1989b Phase IA Archaeological Reconnaissance SurveyWaterbury-Duxbury-Moretown BRF 013-4(10),Washington County, Vermont [VT-WA-35]. Vermont[VT-WA-35]. Consulting Archaeology Program,Department of Anthropology, Report 101. University ofVermont, Burlington.

1991 Archaeological Reconnaissance Survey for Essex-Williston BRZ 1445(13), Essex and Williston, Vermont.Consulting Archaeology Program, Department ofAnthropology, Report 110. University of Vermont,Burlington.

Thomas, P. A. and M. Bourassa1985 Archaeological Evaluation of the ProctorWastewater Treatment Facility; Outflow System, Proctor,Vermont. Department of Anthropology, Report 58.University of Vermont, Burlington.

Thomas, P. A" C. O. Clement and P. Doherty1992 Archaeological Reconnaissance Survey for MiltonBRZ 1448(16) Milton, Vermont. Consulting ArchaeologyProgram, Department of Anthropology. Report 113.University of Vermont, Burlington.

Thomas, P. A. and R. S. Dillon1983 Intensive Archaeological Evaluation for SelectedPrehistoric Sites Within the Highgate FallsHydroelectric Project Area and PrehistoricArchaeological District. Department ofAnthropology, Report 44. University of Vermont,Burlington.

Thomas, P. A. and B. Robinson1980 The John's Bridge Site, VT-FR-69, Area 1, PhaseIII Data Recovery. Department of Anthropology, Report28. University of Vermont, Burlington.

1981 Joseph C. McNeil Generating Station: BorrowArea. Archaeological Evaluation. Department ofAnthropology, Report 31. University of Vermont,Burlington.

Thompson, D. M. and E. A. Bettis III1982 Out of Site, Out of Planning: Assessing andProtecting Cultural Resources in Evolving Landscapes.Contract Abstracts, Vol. 2(3):16-21.

Voight, E. E. and M. J. O'Brien1982 The Use and Misuse of Soils Related Data inMapping and Modeling Past Environments. ContractAbstracts, Vol. 2(3):22-35.

Webb, T., III and D. R. Clark1977 Calibrating Micro-palyontological Data III

Climatic Terms: A Critical Review. In Amerinds andTheir Paleoenvironments in Northeastern North America,edited by W. S. Newman and B. Salwen, pp. 93-118.Annals of the New York Academy of Sciences 288.

Wendlund, W. M. and R. A. Bryson1974 Dating Climatic Episodes of the Holocene.Quaternary Research, Vol. 4:9-24.

PALEOINDIAN PERIOD(11,500-9,000 B.P.)

Regional Context

The waning of the last continental glacier ushered in thefirst human occupation of Vermont and the greaterNortheast about 11,500 years ago. Archaeologists refer tothis time as the Paleoindian period. Most of thePaleoindian sites throughout the Northeast remain undated.Current evidence does suggest that the fluted points foundat such sites as Templeton, Whipple, Bull Brook, Vail,Michaud and Debert (located in Connecticut, NewHampshire, Massachusetts, Maine and Nova Scotia), aswell as most of those recovered at unexcavated site inwestern Vermont, date between 10,000 and 11,000 yearsago (Haynes et a1. 1984). This is a period of mystery, andsites dating to this distant past have the mystique of being"the first" or "the earliest." From Pennsylvania and NewJersey on the south northward to the southern Great Lakesand eastward to Nova Scotia, substantive data about thePaleoindian period can be drawn from fewer than 25 wellexcavated archaeological sites. However, literally thou-sands of Paleoindian artifacts have been recovered, most-ly as isolated surface finds, from hundreds of sitesthroughout the Northeast, and more are being encoun-tered each year (Brennen 1982; Lantz 1984). Thesescattered finds strongly indicate a widespread and per-sistent occupation throughout the region on a far dif-ferent scale than is suggested by the relatively few

44


Figure 2. Paleo indian and Early Archaic period sites cited in the text.

1. Reagen2. Bessette 22. Bessette 33. John's Bridge4. VF-CH-4905. VF-CH-4866. Otter Creek 2

0 10 20 ~o 40 ~ot-3 I I I I

MILES

0 10 jO ,0 41° r r8KILOMETERS

45


extended and detailed studies of Paleoindian sites whichappear in the archaeological literature.

Technology

The most distinctive tool of the Paleoindian period is theClovis-like fluted projectile point. This style of point,first identified in the American southwest in the 1920sand first identified in the Northeast at the Reagen site innorthwestern Vermont, is found at roughly contempora-neous sites throughout North America in a wide variety ofenvironmental settings (Ritchie 1953) (Figure 2). Toolassemblages at some Paleoindian sites are varied andextensive, while at others they are less diversified,Common to many sites are the characteristic flutedpoints, projectile point preforms, leaf-shaped and ovateknives, end scrapers (many with graver spurs), sidescrapers, flake shavers, pieces esquillees (flake wedges),drills, gravers, hammerstones, anvil stones, choppers, andflaking debris from a variety of bedrock sources, almost allof which is of a superior quality for stone tool manufac-ture (Curran 1979; Gramly 1982; Grimes 1979; Spiess1985). The presence of fluted points and spurred endscrapers made from high quality chert has been used toidentify over twenty Paleoindian sites in western Vermont(Loring 1980; Thomas, Dillon and Doherty 1985:74-81,l79-185, 202-204; Dillon, Thomas and Doherty ]985:236-238).

The Reagen site in Highgate, Vermont, is the only sub-stantively reported Paleoindian site in the state. However,interpretation is difficult because all artifacts were pickedup on the surface by collectors after severe wind erosionhad occurred. William Ritchie inventoried 179 stoneartifacts which he arbitrarily divided into four major cat-egories: scrapers, knives, a variety of projectile points, anda few chipped artifacts of indeterminate form. Thirteentalc pendants apparently comprise objects of personaladornment which are unique to the Reagen site (Ritchie1953:251-254), although there are rumors that these pen-dants were planted by collectors to fool their competitors.

The diversity of projectile point styles at the Reagen sitehas fascinated archaeologists for several decades. Of the47 points recovered, point forms include an earred trian-gular type, a small trianguloid type made on broad thinflakes, a probable lanceolate type made by using an obliqueflaking technique, a lanceolate pentagonoid type with verythin fluting on one face or showing horizontal parallelflaking scars across the entire face of the point, a pen-tagonal form, and a stemmed type (Ritchie 1953).

Although similar pentagonal forms have apparently beenencountered at other Paleoindian sites in the Northeast(Snow 1980:130-131), the variability in the Reagenassemblage is unusual and has led archaeologists to inferthat the Reagen site dates to the end of the Paleo indianperiod.

A fInal phase of the Paleo indian period is recognizable in abroad arc extending from the Great Lakes across southernCanada and northern New England and as far east as theGaspee peninsula by the presence of long, very thinlanceolate points which exhibit parallel flake scars fromone edge to the other (Doyle et al. 1986; BenmouyaI1978).To date, the only evidence of this phase in Vermont is sev-eral parallel-flaked specimens from the Otter Creek 2 sitein Rutland County and from a site in the Little Otter Creekdrainage (Ritchie 1969; Doyle et al. 1986:14). Twoobservations are invariably made about Paleoindian arti-fact assemblages. The raw materials used are of veryhigh quality and frequently derive from distant sourceareas. Such patterns exist at Paleo indian sites throughoutNew England where a variety of "exotic" cherts, jaspers,felsites, rhyolites and quartzites are often encountered andwhere source areas are often presumed to lie at distances ofwell over one hundred miles from the sites (Curran 1979;Grimes 1979; Grimes et al. 1984; Gramly 1982, 1984;Spiess 1985; Wilson and Spiess 1990).

At the Reagen site, a wide range of material is also repre-sented, even when the total reported artifact assemblageconsists of only 179 tools and 75 flakes. The most promi-nent raw material among the tools and flaking debris is abanded black and grayish brown chert, with other blackand dark gray cherts being fairly common. Based onmacroscopic characteristics alone, much of this chert ispresumed to be derived from the Champlain Lowland, asare a few artifacts of grayish quartzite. Other raw materi-als consist of whitish, yellowish or reddish rhyolites or fel-sites, traprock, a mottled gray and brown chert, yellow anddark red jasper, green chert and gray chalcedony (Ritchie1953:250-251). Although speculative, source areas for therhyolites, felsites, mottled gray and brown chert, jasperand green chert could extend from eastern Massachusetts,central and eastern New York, and as far south asPennsylvania.

Settlement and Subsistence

In Vermont, Paleoindian artifacts have been recoveredin a variety of geographical settings-from old beachlines and delta tops left behind as the Champlain Sea

46

waned, from the Champlain Lowlands below the maximumlimits of the Champlain Sea, and from high and nearlymountainous terrain in the upper Winooski River water-shed. None have been recovered in modem river bottomsor along the shores of Lake Champlain, but for good rea-son. At that time, modem stream channels and adjacentfloodplains had not yet formed, and Lake Champlain eitherdid not exist or exhibited a lakeshore with a substantiallydifferent configuration during the early stages of its devel-opment.

The presence of "exotic" raw material from distantsource areas at most Paleoindian sites has led manyarchaeologists to conclude that extended family groups orsmall bands moved over wide areas during the courseof the year. Their ability to do this was in part due to theirselection of very high quality stone for tool manufacture,their production of versatile tools and tool kits whichcould be used for a variety of functions, the heavy cura-tion of stone artifacts, and the intensive use of even smallflakes for cutting and scraping. Curation was accomplishedby reworking broken tools into other tools or by creat-ing smaller versions of the same tool. This strategy maypartially explain the diminutive size of the projectilepoints found at the Reagen site (Grimes and Grimes 1985;Lowery 1989:161; Snow 1980:129; Ritchie 1953). Analternative model proposes that the movement of familygroups was less wide ranging and that "exotic" stone wasprocured by specialized task groups who traveled to thequarries and brought necessary materials back to the hometerritory (Wilson and Spiess 1990). So little is knownabout the characteristics and evolution of Paleoindian set-tlement patterns that any definition of site types must betentative at best. Habitation sites are the most readilyobserved. Because there is a substantial range in sizerange in size and artifact density among sites, however,large base camps and small residential camps are provi-sionally offered as site categories. The category of smallresidential camp might include such sites as Reagen innorthwestern Vermont.

One pattern is characteristic of both types of sites: arti-facts are typically concentrated in distinct clusters with-in a larger site area. This distributional pattern wasnoted as early as 1950 by William Ritchie at the Reagensite and has been confirmed at other sites since then(Ritchie 1953:249). For example, at the Whipple site locat-ed on a sandy terrace above the Ashuelot River in NewHampshire, Curran identified three distinct activity areas of6-8 m diameter spaced 60-100 m apart. At the Bull BrookI in eastern Massachusetts, approximately 42 activity loci


of 5-10 m diameter were identified within an area of 8,000m2 (Grimes and Grimes 1985; Spiess et al. 1985). Giventhe fairly harsh climate which prevailed, such artifactconcentrations could represent individual shelters withinwhich people carried out a variety of activities. To date,however, the remnants of an actual shelter or househave not been identified. Clearly, much remains to belearned about the functional nature of Paleoindian sites,how various types of sites were integrated into a largersettlement system, and how such patterns may havechanged through time.

The apparent evidence for long distance movements ofpeople on an annual basis and the identification ofPaleoindian sites in a variety of environmental settingshave led to extended discussions and attempted expla-nations of Paleoindian subsistence patterns. What typesof food resources were utilized? Were Paleoindians "spe-cialists" or "generalists?" How did they organize?

In the western part of the United States, "Clovis adaptivepatterns are heavily skewed towards task-specific activi-ties of mammoth killing and processing" (Lahren1974:148). This same theme, which tends to viewPaleoindians as "specialists" who focused their attentionson one or two large game animals, was picked up in theEast by various archaeologists (Ritchie 1957; Funk 1970,1972; Dragoo 1979). This view has been challengedby other archaeologists who hypothesize that Paleoindianswere "generalists" who were more likely to have utilizeda wide variety of food resources rather than focusing onone specific type of game (Dincauze and Curran 1983;Curran 1984; Spiess, Curran and Grimes 1985:156).Today, the debate tends to be one of degree rather thankind. Paleoindians may have utilized both strategies,depending upon the season or other environmental factors.Faunal remains from Northeastern Paleoindian sites arerare, but where they have survived, a mixed diet is sug-gested, with caribou, beaver, fish, large bird, and charredhawthorn pits having survived at various sites (Spiess,Curran and Grimes 1985:146-148; Funk et al. 1969;Kooper, Funk and Dumont 1980; Eisenberg 1978).Until plant and animal remains are identified from a greatmany more early sites, characterizations of Paleoindiansubsistence practices will remain problematic.

Ideology

Given the great antiquity of Paleoindian sites and theobvious lack of any modem cultural analogs which mightbe used to infer the nature of the spiritual beliefs of these

47


people, it seems safe only to say that they possessed a com-plex set of beliefs which we can never hope to under-stand. From widely scattered sites in North America, wesee only fleeting glimpses of such beliefs which relateto the treatment of their dead comrades and relatives. Atthe Anzick site in Montana, for example, the first knownPaleoindian burial was identified in a rock shelter in 1973.'The combination of ceremonial breaking [of stone andbone tools] and red ocher covering is probably the earliestevidence of New World religion" (Lahren and Bonnichsen1974).

Closer to Vermont, in southwestern Ontario, Deller andEllis (1984) make a convincing case for the presence ofa cremation burial at the Crowfield site, despite the absenceof identifiable human bone. A pit measuring approxi-mately 1.5 m wide and roughly 45 em deep was foundto contain over 1,400 fire-shattered artifacts, includingfluted points and other typical Paleoindian tools. Theseauthors further note that a similar cremation was encoun-tered at the late Paleoindian Renier site in Wisconsin,and that, based on stylistic similarities of points found atthe Reagen site in northwestern Vermont, the Crowfieldand Reagen sites may date to approximately the same peri-od (Deller and Ellis 1984:50).

Although these glimpses into the distant past are tantaliz-ing, it seems fair to say that human burials dating to thePaleoindian period will rarely be encountered in thefuture and that Paleoindian ideology is likely to remainfirmly in the realm of speculation.

Paleoindian Period: References

Benmouyal, J.1987 Des Paleoindiens aux Iroquoiens en Gaspesie: SixMille Ans d'Histoire. Ministere des Affaires Culturelles,Direction de I'Est du Quebec, Quebec.

Brennen, L. A.1982 A Compilation of Fluted Points of Eastern NorthAmerica by Count and Distribution: An AENA Project.Archaeology of Eastern North America, Vol. 10:27-46.

Curran, M. L.1979 Studying Human Adaptation at a Palco-Indian Site:A Preliminary Report. In Ecological Anthropology of theMiddle Connecticut River Valley, edited by R. Paynter, pp.14-26. University of Massachusetts Research Report 18.

1984 The Whipple Site and Paleoindian Tool Assemblage

Variation: A Comparison of Intrasite Structuring.Archaeology of Eastern North America, Vol. 12:5-40.

Deller, D. B. and C. J. Ellis1984 Crowfield: A Preliminary Report on a ProbablePaleo-Indian Cremation in South-western Ontario.Archaeology of Eastern North America, Vol. 12:41-71.

Dillon, R. S., P. A. Thomas and P. Doherty1985 Archaeological Reconnaissance Survey and SiteEvaluation Study for the Yellow-Blue-Yellow Alignment,Chittenden County Circum-ferential Highway, 1984-1985.Volume V: Technical Data for Sampling Areas andArchaeological Sites. Department of Anthropology,University of Vermont, Report No. 65, pp. 236-238.

Dincauze, D. and M. L. CUrran1983 Paleoindians As Generalists: An EcologicalPerspective. Society for American Archaeology,Abstracts, Annual Meeting, April 27-30.

Doyle, R. A., N. D. Hamilton, J. B. Petersen and D. Sanger1985 Late Paleo- Indian Remains from Maine andTheir Correlations in Northeastern Prehistory.Archaeology of Eastern North America, Vol. 13:1-34.

Dragoo, C. W.1979 The Proboscidians and Man. Archaeology ofEastern North America, Vol. 7:180-213.

Eisenburg, L.1978 Paleo-Indian Settlement Pattern in the Hudsonand Delaware River Drainages. Man in the Northeast,Occasional Publications in Northeastern Anthropology,Vol. 4.

Funk, R. E.1970 Caribou and Paleo-Indians in New York State: APresumed Association. American Journal of Science, Vol.268:181-186.

1972 Early Man in the Northeast and the Late-glacialEnvironment. Man in the Northeast, Vol. 4:7-39.

Funk, R. E., G. R. Walters, W. F. Ehlers, J. E. Guilday andO. O. Connally1969 The Archaeology of Dutchess Quarry Cave.Orange County, New York. Pennsylvania Archaeologist,Vol. 39 (1-4):7-22.

48

Gramly, R. M.1982 The Vail Site: A Palaeo-Indian Encampment inMaine. Bulletin of the Buffalo Society of NaturalSciences, Vol. 30. Buffalo, New York.

1984 Kill Sites, Killing Grounds and Fluted Points atthe Vail Site. Archaeology of Eastern N0/1h America, Vol.12:110-121.

Grimes, J. R.1979 A New Look at Bull Brook. Anthropology, Vol.3:109-130.

Grimes, 1. R., W. Eldridge, B. G. Grimes, A. Vaccaro, F.Vaccaro, 1. Vaccaro, N. Vaccaro and A. Orsini1984 Bull Brook II. Archaeology of Eastern NorthAmerica, Vol. 12:159-183.

Grimes, 1. R. and B. G. Grimes1985 Flak:eshavers: Morphometric, Functional and Life-cycle Analyses of a Paleoindian Unifacial Tool Class.Archaeology of Eastern North America, Vol. 13:35-57.

Haynes, C. V., D. .T.Donahue, A. 1. T. lull and T. H. Zabel1984 Application of Accelerator Dating to Fluted PointPaleoindian Sites. Archaeology of Eastern North America,Vol. 12:184-191.

Kopper, 1. S., R. E. Funk and L. Dumont1980 Additional Paleo-Indian and Archaic Materialsfrom Dutchess Quarry Cave Area, Orange County, NewYork. Archaeology of Eastern North America, Vol. 8: 125-137.

Lahren, L. and R. Bonnichsen1974 Bone Foreshafts from a Clovis BurialSouthwestern Montana. Science, Vol. 186:147-150.

Lantz, S. W.1984 Distribution of Paleo-Indian Projectile Points andTools from Western Pennsylvania: Implications forRegional Differences. Archaeology of Eastern NorthAmerica, Vol. 12:210-221.

Loring, S.1980 Paleo-Indian Hunters and the Champlain Sea: APresumed Association. Man in the Northeast, Vol. 19:15-41.

Lowery, D.1989 The Paw Paw Cove Paleoindian Site Complex,


Talbot County, Maryland. Archaeology of Eastern NorthAmerica, Vol. 17:143-164.

Ritchie, W. A.1953 A Probable Paleoindian Site in Vermont.American Antiquity, Vol. 18:249-258.

1957 Traces of Early Man in the Northeast. New YorkState Museum and Science Service Bulletin 358. Albany,New York.

1969 The Archaeology of New York State [revised].Garden City: Natural History Press.

Snow, D. R.1980 The Archaeology of New England. New York:Academic Press.

Spiess, A. E.1985 The Michaud Site (23-12): A New Major Fluted-Point Paleo-Indian Site in Auburn, Maine. The MaineArchaeological Society Bulletin, Vol. 25(2):38-42.

Spiess, A. E., M. L. Curran and 1.R. Grimes1985 Caribou (Rangifer tarandus L) Bones from NewEngland Palaeoindian Sites. North AmericanArchaeologist, Vol. 6(2):145-159.

Thomas, P. A., R. S. Dillon and P. Doherty1985 Archaeological Reconnaissance Survey for theChittenden County Circumferential Highway. VolumeII: Site Descriptions and Technical Information.Department of Anthropology, University of Vermont,Report No. 52, pp. 74-81,179-185,202-204.

In Wilson, D. and A. Spiess1990 Study Unit 1: Fluted Point Paleoindian. MaineArchaeological Society Bulletin, Vol. 30(1):15-31.

EARLY ARCHAIC PERIOD(9000-7500 B.P.)

Regional Context

The identification of Early Archaic sites throughout NewYork, New England and the eastern Canadian provinceshas been largely determined by the presence of a series ofprojectile point types first recovered from stratified allu-vial sequences in the Southeast. As defined in the

49


Southeast, the basic typological sequence runs from earlyHardaway-Dalton points, to Kirk Comer Notched, St.Albans Side-Notched, and Le Croy Bifurcate Stemmed,followed by Kanawha Stemmed and Kirk Stemmed(Broyles 1966, 1971; Chapman 1980; Coe 1964). Inthe Northeast, recognition has come slowly, and priorto 1966 "no Archaic assemblages in Pennsylvania, NewYork, New England, and the Upper Great Lakes could beconfidently assigned an age as great as 3000 B.C." [5000B.P.] (Funk 1978:20).

Among New England archaeologists, perceptions aboutthe Early Archaic period have been evolving. The ideaof a cultural hiatus in the Northeast between thePaleoindian period and a later Archaic period postdating6000 B.P. was widely held (Fitting 1968; Ritchie 1969),but is now considered inaccurate (Robinson, Petersen andRobinson 1992). There is no clear consensus, however,on how to interpret the evidence of widely scattered EarlyArchaic period sites which have been defined primarily bythe presence of the more easily identifiable variety ofbifurcate base points in the New York, New England andsouthern Ontario area. Low site densities when comparedto site densities of later periods have generally beentaken to mean a low human population density through-out the region during the Early Archaic and subsequentMiddle Archaic periods. The potential for a substantialsampling bias has been raised repeatedly (Dincauze andMulholland 1977; Petersen et a1. 1986; Thomas 1992),and Northeastern archaeologists may be well advised toconsider the lessons learned in the Southeast. AsChapman (1980:129) noted a decade ago, "The per-spective from eastern Tennessee concerning the Early andMiddle Archaic periods suggests a greater population den-sity than previously thought. In just over twenty years, wehave moved from the position of little evidence for theArchaic period to the recognition of extensive occupationof the active alluvial terraces adjacent to the river margins.Upland ridge sites and deflated hill tops are now viewedas perhaps marginal and probably secondary to the floodplain sites."

As a result of archaeological projects conducted in NewEngland during the past decade, it is clear that the existenceof deeply buried Early and Middle Archaic period sites onriverine terraces is not unique to the Southeast. Culturalmaterial recovered from the base of a number of flood-plain sites are clearly attributable to Early Archaic periodoccupations dating between 9000 and 7500 B.P. Theseinclude such sites as the Walnut Street site located onthe Connecticut River just south of the Vermont border

in Gill, Massachusetts, Wadleigh Falls located on theLamprey River in southeastern New Hampshire, theShallow and the Cape and Wood Island sites located onthe Piscataquis and Androscoggin Rivers in Maine, theEddy site located at Arnoskeag Falls on the MerrimacRiver in New Hampshire, and Bessette 3 and Bessette 2located on the Missisquoi River in northwestern Vermont,and dated to 7970 ± 270 and 7730 ± 70 B.P., respectively(Curran and Thomas 1979; Robinson, Petersen andRobinson 1992; Figure 2).

Technology

Early Archaic period sites are likely to contain one or sev-eral varieties of comer-notched, stemmed and bifurcatebase projectile points which are usually referred to bynames used for their Southeastern analogs. These pointshave been found in association with bifaces in theprocess of being reduced to finished points, end-scrapers,side-scrapers, utilized flakes, and far less often with suchitems as hammerstones, milling slabs, abraiding stones, pit-ted stones, notched pebble netsinkers, and ovate chop-pers with flaked edges. Expediency tools appear todominate all artifact assemblages, even at large EarlyArchaic sites in the Southeast. As Chapman (1980: 127)has noted for eastern Tennessee, "Within the Early andMiddle Archaic lithic assemblages, it is important to notethe large number of ad hoc implements representing onlyslightly modified flakes and blades; formalized chippedston~, tool types are relatively infrequent."

Both parallels and differences occur in artifact assem-blages from excavated Early Archaic period sites inVermont. In 1980, the small, single component John'sBridge site was encountered on a bedrock-defended ter-race above the Missisquoi River in Swanton, Vermont(Thomas and Robinson 1980). Three radiocarbon sam-ples (7780 ± 225, 8240 ± 240, 8340 ± 245 B.P.) derivedfrom a deep pit feature yielded a mean date of approxi-mately 8100 B.P. Rather than some variety of bifurcatebase point, the John's Bridge site contained a sample ofeleven whole or partial points which are triangular orslightly ovate in outline, comer-notched, with straight orslightly indented bases which have usually been ground.Because this point exhibited a morphology unlike otherrecognized contemporaneous point forms in New Yorkand New England, it was assigned the name SwantonComer-Notched (Plate l ). Since that time, such pointshave been recognized at 13 additional sites in theChamplain Basin (Thomas and Robinson 1980; Thomas1992; Frink 1989), and analogs may exist in western

50

Plate 1. Swanton Corner-Notched points from the John'sBridge Site, Swanton, Vermont.

Maine (Spiess et a1. 1983), at the Harrisena site (Snow1977) located on the divide between the Hudson River andLake George which drains north into Lake Champlain, andfrom Lake St. Francis on the upper St. Lawrence River(Plate 2).

In addition to the 23,165 chert (52.6%), quartzite (20.6%)and quartz (26.8%) flakes recovered in a large work-shop/shelter area, artifacts related to the manufacture ofpoints include chert and quartzite quarry blanks, broadovate preforms, biface fragments, and an abraiding stonewhich may have been utilized to prepare striking platformsduring various stages of biface reduction. In addition, awide variety of cutting and scraping tools were recoveredfrom the 33.5 m2 area which was excavated. These includebifacial flake tools, including one perforator, several vari-eties of scrapers produced from thick chunks of chert orfrom thick flakes, utilized flakes which appear to havebeen employed as the primary cutting tools at the site,"nibbled" flakes, and large tabular blades. Six small nod-ules of graphite were also recovered, the largest of whichwas clearly ground on several sides. This assemblage con-tains nearly the full range of formalized and expedien-


Plate 2. Swanton Corner-Notched points from LeichesterFlats: VT-AD-64 (upper left and center); and from theEwing (upper right) and Auclair (others) sites onShelburne Pond. Photo by James B. Petersen.

cy tools which might be expected at Early Archaic peri-od sites. Evidence of both surface hearths (without fire-cracked rock) and several deep pits was also recovered.

Although the association is less clear, a drilled, rec-tangular-shaped bannerstone or atlatl made from bandedslate and a notched pebble which may have been used asa netsinker were recovered in close proximity toSwanton coner-notched points during intensive surfacesurveys along the Lamoille and Missisquoi Rivers,respectively. In each instance, the site was small andappeared to be single component (Thomas et a1. 1980:93;Thomas et a1. 1981:139).

Partial data recovery has been undertaken at two EarlyArchaic sites in Highgate located on a high alluvial terracealong the Missisquoi River. Only 12 square meters havebeen excavated at Bessette 3. The recovery of portions oftwo bifurcate base points, less than 200 chert flakes and ashallow hearth would seem to indicate the presence of asmall residential camp. A radiocarbon date of 7970 ± 270

51


B.P. was obtained from the hearth (Thomas 1992).

Bessette 2 is located some 130 m upstream. Based onthe geomorphology and a radiocarbon date of 7730 ± 70B.P. from charcoal stratified above a distinct layer of arti-facts, the occupation of Bessette 2 must be older, prob-ably by several hundred years. The four projectile pointsrecovered from an area of 67 m2 are difficult to place typo-logically. Two of the points have nearly parallel sides,small side notches and ground bases. They are similar toone notched point with a ground base recovered at theJohn's Bridge site, and thus may be fairly contemporane-ous. In addition to the four points and a 15 m2 workshopwhich contained over 10,000 chert flakes and biface frag-ments, the assemblage consisted predominantly of ovoidbifaces, flake knives, scrapers, large and fairly crude cut-ting and chopping tools, and two tabular knives or chopperwith flaked edges which are similar to those recovered atthe John's Bridge site downstream (Thomas 1992). Atthe John's Bridge site, as well as at Bessette 2 and 3, botha bifacial and unifacial lithic technology are well repre-sented.

At several recently excavated sites in Vermont, includ-ing VT-CH-486 and VT-CH-490 located along IndianBrook in Essex, and in northern New England, unifacialscrapers made from angular pieces of quartz and quartziteand flakes with slightly modified edges were the domi-nant artifacts recovered; finished projectile points or evenevidence of bifacial reduction are limited (Robinson,Petersen and Robinson 1992).

Comparisons of the artifact assemblages from these siteslead to several inferences. First, considerable variationmay be expected from site to site depending upon thelength of residence and the specific types of activitieswhich the occupants carried out. Second, there is suffi-cient variability in projectile point forms so that site identi-fication based on point morphology alone may notalways prove reliable. Finally, virtually all of the lithicmaterial recovered is thought to be derived from localsources of chert, quartzite or quartz. Except for the pres-ence of a single piece of what appears to be Onondagachert from western New York at the John's Bridge site, theuse of "exotic" lithic materials from distant source areasfor tool manufacture so characteristic of the earlierPaleoindian period is uncommon. This fact leads to theinference that people had settled into Vermont by thistime and knew where to easily find workable stone.


In Vermont, Early Archaic period sites have been identi-fied in a variety of geographical settings within theChamplain Lowland of western Vermont, in the VermontValley situated between the Green and Taconic Mountainsin southwestern Vermont and in the valley bottom alongthe Connecticut River in southeastern Vermont (Thomas1992). It seems likely that by the end of the Early Archaicperiod bands were settling into the region and mayhave developed home territories. Movement within a ter-ritory could be fairly erratic or more highly scheduled on aseasonal or other basis depending upon resource pre-dictability. Habitation sites probably include both basecamps situated in high resource areas and smaller resi-dential camps located in the surrounding territory.

One pattern is very likely to be characteristic of bothtypes of sites: artifacts will be typically concentrated indiscrete focal areas. Given the severe weather conditionsduring some seasons of the year, some focused artifact con-centrations could reflect the presence of individual shel-ters within which people carried out a variety of activities.One such shelter was inferred based on the distribution ofartifacts at the John's Bridge site (Thomas and Robinson1980). To date, however, the physical remains of anactual shelter or househave not been encountered.

The wide distribution of Early Archaic sites in Vermontdating to approximately 8000 B.P. supports the inter-pretation of a resident population which had developedstrategies to exploit a wide range of food resources in anumber of settings. Swanton comer-notched points have,for example, been identified at fourteen sites in theChamplain Basin. These sites are situated on riverine ter-races along Otter Creek and the Lamoille and MissisquoiRivers, on the outlet of Shelburne Pond in the Winooskiwatershed, and on the high, sandy, outwash delta plainsabove the Lamoille and Winooski Rivers in Milton andColchester. Sites containing bifurcate base points exhib-it an even wider distribution, ranging from Bennington inthe south to Highgate in the north. Site locations current-ly include one site on the shore of Lake Champlain,three sites on alluvial terraces adjacent to the MissisquoiRiver, Walloomsac River and Otter Creek, five sites onbedrock islands surrounded by wetlands along OtterCreek, four sites on the shore of Shelburne Pond, LakeIroquois, Bristol Pond and Fletcher Beaver Pond, and fourlocations on rolling lake plains in the Indian Brook water-shed and in the foothills of the Green Mountains (Thomas1992). Site locations strongly suggest the use of at least

52

riverine, lacustrine/wetland and lowland forest resources.

Faunal remains from Early Archaic sites in New Englandare rare and generally consist of burned fragments ofbone which have been heavily eroded by natural weather-ing in the soil. Where they have survived, however, amixed diet is suggested. Faunal remains are particularlysparse at Early Archaic sites in Vermont. At the John'sBridge site, for example, 71 small fragments of calcinedmammal bone and one fragment of antler cannot be iden-tified as to species, but of the six pieces of fish bone, twoare pectoral spines from either a bullhead or channel cat-fish. A fragment of charred nut shell, possibly hazelnut,was recovered in a deep pit (Thomas and Robinson1980:51-54). Hearths were identified at both Bessette 2and 3, located on a high riverine terrace farther upstreamalong the Missisquoi River. However, only small frag-ments of unidentifiable mammal bone were recovered,even after intensive flotation of feature soils. Distinctcultural features containing faunal remains were not iden-tified at VT-CH-490 and VT-CH-486. All other EarlyArchaic sites in Vermont have been identified by eithersurface finds of Swanton corner-notched or bifurcatebase points or at unstratified, multicomponent sites wherefaunal remains cannot be clearly associated with a specificperiod of occupation.

Fundamental questions remain about temporal sequences,the representativeness of current site data, and the natureand variability of the early Holocene environment.There no longer seems to be a question of a continuousoccupation of New England following the Paleoindianperiod. What remains, however, is a very poor under-standing of the factors which may affect site discovery,the cultural systems which were active during this period,and the complex natural environment to which people hadadapted. Detailed analyses and descriptions of EarlyArchaic period tool assemblages from other sites areclearly needed in order to increase the possibility of siterecognition, evaluate the extent of cultural affinity and/orcontemporaneity of identified sites throughout Vermontand the Northeast, and to address the issue of potential sitefunction.

Much remains to be learned about the functional and sea-sonal nature of Early Archaic sites, how various types ofsites were integrated into a larger settlement system, andhow such patterns may have changed through time. Onething is clear. The presence of Early Archaic sites in deepalluvial deposits along the state's major rivers, in areascurrently submerged beneath Lake Champlain or In


environments which are not usually surveyed, as well asthe highly focused distribution of artifacts within indi-vidual sites even where soils are fairly shallow means thatEarly Archaic sites will continue to be very difficult tolocate.

Early Archaic Period: References

Broyles, B. J.1966 Preliminary Report: The St. Albans Site (46Ka27),Kanawha County, West Virginia. The West VirginiaArchaeologist, Vol. 19.

1971 Second Preliminary Report: The St. Albans Site,Kanawha County, West Virginia. West VirginiaGeological and Economic Survey, Report ofArchaeological Investigation 3.

Chapman, J.1980 The Early and Middle Archaic Periods: APerspective from Eastern Tennessee. In Proceedings ofthe Conference on Northeastern Archaeology, edited by J.A. Moore, pp. 123-132. Research Report Number 19,Department of Anthropology, University of Massachusetts,Amherst.

Coe, J. L.1964 The Formative Cultures of the Carolina Piedmont.Transactions of the American Philosophical Society, Vol.54 (5), new series: 1-130.

Curran, M. L. and P. A. Thomas1979 Phase III Data Recovery: Wastewater TreatmentSystem in the Riverside Archaeological District of Gill,Massachusetts. Department of Anthropology, Report 19.University of Vermont, Burlington.

Dincauze, D. and M. T. Mulholland1977 Early and Middle Archaic Site Distributions andHabitats in Southern New England. In Amerinds and TheirPaleoenvironments in Northeastern North America, edit-ed by W. S. Newman and B. Salwen, pp. 439-456. Annalsof the New York Academy of Sciences 228.

Fitting, 1. E.1968 Environmental Potential and the Post-glacialReadaptation in Eastern North America. AmericanAntiquity, Vol. 33 (4):441-445.

Frink, D.1989 The Blue Heron Site-VT-CH-363: An Early

53


Archaic Food Processing Site. Archaeology ConsultingTeam, Essex Junction, Vermont.

Funk, R. E.1978 Post-Pleistocene Adaptations. In Northeast, editedby B. Trigger, pp. 16-27. Handbook of NorthAmerican Indians, Vol. 15, William G. Sturtevant, gen-eral editor. Smithsonian Institution, Washington, D.C.

Petersen, J. B., N. D. Hamilton, D. E. Putnam, A. E.Spiess, R. Stuckenrath, C. A. Thayer and J. J. Wolford1986 The Piscataquis Archaeological Project: A LatePleistocene Occupational Sequence in Northern NewEngland. Archaeology of Eastern North America, Vol.14:1-18.

Ritchie, W. A.1969 The Archaeology of New York State [revised edition].Garden City: Natural History Press.

Robinson, B. S., J. B. Petersen and A. K. Robinson1992 Early Holocene Occupation in Northern NewEngland. Occasional Publications in Maine Archaeology,Number Nine.

Snow, D.1977 The Archaic of the Lake George Region. InAmerinds and Their Paleo-environments in NortheasternNorth America, edited by W. S. Newman and B. Salwen,pp. 316-332. Annals of the New York Academy ofSciences, 228.

Spiess, A. E., B. J. Bourque and R. M. Gramly1983 Early and Middle Archaic Site Distribution inWestern Maine. North American Archaeologist, Vol.4:225-244.

Thomas, P. A.1992 The Early and Middle Archaic Periods AsRepresented in Western Vermont. In Early HoloceneOccupation in Northern New England. OccasionalPublications in Maine Archaeology, Number Nine.Augusta, Maine.

Thomas, P. A. and B. S. Robinson1980 The John's Bridge Site: VT-FR-69. Department ofAnthropology, Report 28. Reprinted by the VermontArchaeological Society, Burlington.

Thomas, P. A., W. A. Bayreuther, G. Campoli and P.Doherty

1980 Phase I Archaeological Assessment for the EastGeorgia Hydroelectric Project. Report No. 19,Department of Anthropology, University of Vermont.

Thomas, P. A., G. Campoli, P. Doherty, and J. Petersen1981 Phase I Archaeological Assessment for theMissisquoi Hydroelectric Project: Sheldon Springs,North Sheldon, Enosburg Falls, North Troy and JayBranch Elements. Report No. 34, Department ofAnthropology, University of Vermont.

MIDDLE ARCHAIC PERIOD(7500-6000 B.P.)

Regional Context

The identification of Middle Archaic sites throughoutsoutheastern New York, New England and the easternCanadian maritimeprovinces has been largely determinedby the presence of three projectile point types firstdescribed by Dincauze from the base of the stratifiedNeville site on the Merrimack River in Manchester, NewHampshire. Like those in the Early Archaic sequence,these points, designated Neville, Neville variant and Starkpoints, have their analogs in the Southeast where they aregenerally referred to as Stanly and Morrow Mountain II(Coe 1964; Dincauze 1971).

At the Neville site, dates of7,015 ± 160,7210 ± 140,7650± 400, and 7740 ± 280 B.P. recovered from Stratum 5which contained the Neville and Stark points provide arough age bracket for this complex. At the Neville site,as well as at the stratified WMECO site on theConnecticut River in Gill, Massachusetts, a layer con-taining virtually no artifacts marks a pronounced strati-graphic break above the Middle Archaic horizon. Thisstratigraphic hiatus is dated by two radiocarbon sample,5910 ± 180 and 6060 ± 130 B.P., at the Neville site. Atboth sites, Neville and Stark points were found below;Otter Creek, Vosburg and Brewerton points dating to theLate Archaic period were recovered above (Dincauze1976; Thomas 1980). Radiocarbon dates from featureswith clearly associated Neville or Stark points are lackingfrom other sites in New England.

For many New England archaeologists. substantialincreases in the number of Neville and Stark points in col-lections and at excavated sites relative to Early Archaic

54

types signal a significant population increase throughoutthe region (Dincauze and Mulholland 1977; Spiess,Bourque and Gramly 1983; Starbuck 1982; Cassedy 1983;Yesner, Hamilton and Doyle 1983). With a comparableresource base and complex of rivers, lakes, ponds and wet-lands in the Champlain Lowland and in the ConnecticutValley to the east, it seems highly unlikely that thearchaeological record of the Middle Archaic period inVermont would not in some ways mirror the trend forincreased site density, as well as the settlement and sub-sistence patterns documented at sites in Massachusetts,New Hampshire and Maine. To date, however, suchsites have gone virtually undocumented in the entire stateand in areas to the west - a pattern which has per-plexed archaeologists working in the upper St. Lawrenceand Great Lakes region for over a decade.

Technology

As noted previously, Middle Archaic period sites are like-ly to contain one or several varieties of stemmed projectilepoints which are usually referred as Neville, Neville vari-ant and Stark. In addition to the projectile points, arti-fact assemblages from various sites include preforms,unhafted flake scrapers, tiny quartz scrapers, flakeknives, perforators, winged atlatl weights, bifacialchipped knives, plummets, adzes, gouges, full groovedaxes, pebble and cobble hammers and heavy flaked chop-pers (Dincauze 1976). As during the preceding EarlyArchaic period, expediency tools appear to dominate mostif not all Middle Archaic period artifact assemblages.

In Vermont, the occurrence of Neville and Stark points israre, and the number of associated sites is far lower thanfor the preceding Early Archaic period. After an extendedreview of over 5,000 projectile points in the FlemingMuseum collection at the University of Vermont, onlythree Neville points were identified, all from ChittendenCounty; no Stark points were located (Thomas 1992). Areview of the state site inventory led to the recognition ofonly six possible Middle Archaic sites. Finally, in over100 surveys conducted throughout the state by UVM'sConsulting Archaeology Program since 1978, no Nevilleand only one Stark projectile point has ever been iden-tified. The latter point was recovered as an isolated find onan old alluvial terrace next to Otter Creek in Rutland(Thomas et a1. 1983:66, Appendix B). Based on this evi-dence, or lack of it, can we draw the conclusion that fewpeople were living in Vermont between 7500 and 6000years B.P.? It seems highly unlikely, and resolving thisquestion is a primary research goal.


The low frequency of recognized Middle Archaic sites maystem from the fact that Neville and Stark points are not typ-ical forms in the artifact assemblages of this period inthe Champlain Basin. The weak representation of Nevilleand Stark points in the Champlain Lowland, south intothe upper Hudson River drainage. as well as to the westin the upper St. Lawrence and around Lake Ontario, sug-gests the probability that other Middle Archaic pointstyles, or even a different technology, existed in this entireregion. If such styles or technologies can be identified, itmay imply that people living in the Champlain Basin andother communities located throughout the upper St.Lawrence-Great Lakes drainage at this time had closer cul-tural tie with one another than they did with the Atlanticcoastal communities.


Recognition of Middle Archaic period sites in Vermont isso limited at this time that little can be said about settle-ment patterns. Furthermore. no subsistence data havebeen recovered from any Middle Archaic period site inVermont. Site location provides the best clue toresource exploitation at this time. One Neville point hasbeen reported near the outflow for Shelburne Pond(Petersen et al. 1985:59,65). Two possible Neville pointshave been reported from a site on the shore of LakeChamplain in Ferrisburg. Two Neville points were col-lected by James Petersen on a high terrace adjacent to theLeicester River in the Otter Creek watershed. In south-western Vermont, one Neville point is present in theBenford collection of artifacts from Lake Bomoseen and asecond is present in the Lathrop collection, presumablyfrom a site on the Batten Kill in Sunderland. In southeast-ern Vermont, one Stark point is reported in the Coane col-lection from the Salmon Hole along the upper WestRiver which drains into the Connecticut River. This sitedistribution suggests a strong riverine, pond and lake ori-entation with a mixed economy based on fishing, huntingand gathering. Faunal remains from Middle Archaic sitesin New England are fairly unconunon. However, wherecomponents are buried in deep alluvium and whereflotation techniques have been intensively employed.faunal remains indicate a very diversified diet. In somecases, the variety of species recovered from MiddleArchaic period sites is greater than that recovered fromsites of all later periods. At the WMECO site on theConnecticut River. for example, bones from shad oralewife, muskrat. bird, turtle and snake were recovered(Thomas 1980). At Wadleigh Falls on the Lamprey River

55


in southern New Hampshire, turtle and snake bones dom-inate the assemblage, but beaver, shad, muskrat, deer, catand unidentified bird are also present, while at the Lundsite, near the outlet of Lake Cobbosseecontee in Maine,beaver and muskrat dominate the assemblage, with lowoccurrences of bear, deer, fisher, otter, mink, fox, loon,cormorant, gull, snake, turtle, shad, trout and sucker(Spiess 1992; Spiess, Bourque and Grarnly 1983). Basedon a number of lines of evidence, it is apparent thatgroups during the Middle Archaic period had settled intothe region, that a number of physiographic andmicro-environmental settings were utilized, and thatsubsistence was based on a wide range of food resources.A great deal of variability may actually exist among sitesat this time. In the final analysis, such factors as sea-sonality of site occupation, the environmental char-acteristics of the territory surrounding any specificsite, the chronological placement of a site within this1,500 year period; and the specific types of strategieswhich people used to hunt, fish, gather, consume andpossibly store various food items, as well as the site spe-cific conditions of organic preservation, will have agreat deal to do with the types of food resources whichmight be represented.

Ideology

Given the antiquity of Middle Archaic sites, it seemshighly unlikely that archaeologists will ever be able toinfer the nature of the spiritual beliefs of these people.Even evidence of burial ceremonialism which might reflectbeliefs related to the treatment of deceased friends and rel-atives is currently lacking in all non-coastal areas of theentire Northeast.

Middle Archaic Period: References

Cassedy, D. F.1983 Middle Archaic Stemmed Points in Eastern NorthAmerica. Paper presented at the 50th annual meeting ofthe Eastern States Archaeological Federation, Salem,Massachusetts.

Coe, 1. L.1964 The Formative Cultures of the Carolina Piedmont.Transactions 0/ the American Philosophical Society, Vol.54 (5), new series: 1-130.

Dincauze, D. F.1971 An Archaic Sequence for Southern New England.American Antiquity, Vol. 36(2): 194-198.

1976 The Neville Site: 8,000 Years at Amoskeag.Peabody Museum Monograph 4. Cambridge,Massachusetts.

Dincauze, D. and M. T. Mulholland1977 Early and Middle Archaic Site Distributions andHabitats in Southern New England. In Amerinds and TheirPaleoenvironments in Northeastern North America, edit-ed by W. S. Newman and B. Salwen, pp. 439-456. Annalsof the New York Academy of Sciences 228.

Petersen, 1. B., 1. A. Walford, N. D. Hamilton, L. LaBarand M. Heckenberger1985 Archaeological Investigations in the ShelburnePond Locality, Chittenden County, Vermont. Annals of theCarnegie Museum, Vol. 54(3):23-75.

Spiess, A. E.1992 Archaic Period Subsistence in New England and theAtlantic Provinces. In Early Holocene Occupation inNorthern New England, edited by B. S. Robinson, 1. B.Petersen and A. K. Robinson, pp. 163-185. OccasionalPublications in Maine Archaeology, Number Nine.Augusta, Maine.

Spiess, A. E., B. 1. Bourque and R. M. Grarnly1983 Early and Middle Archaic Site Distribution inWestern Maine. North American Archaeologist, Vol.4:225-244.

Starbuck, D. R.1982 A Middle Archaic Site [in} Belmont, NewHampshire. Report prepared for the New HampshireDepartment of Public Works and Highways byArchaeological Research Services, Durham, NewHampshire.

Thomas, P. A.1980 The Riverside District, The WMECO Site, andSuggestions for Archaeological Modeling. In Early andMiddle Archaic Culture in the Northeast, edited by D. R.Starbuck and C. E. Bolian, pp. 73-96. OccasionalPublications in Northeastern Anthropology, Vol. 7.

1992 The Early and Middle Archaic Periods AsRepresented in Western Vermont. In Early HoloceneOccupation in Northern New England, edited by B. S.Robinson, 1. B. Petersen and A. K. Robinson, pp. 187-203. Occasional Publications in Maine Archaeology,Number Nine. Augusta, Maine.

56

Thomas, P. A., P. Doherty, R. S. Dillon and R. Reinke1983 Archaeological Reconnaissance and Intensive SiteEvaluation Report for the Proposed Route 4 Bypass: WestRutland-Rutland F020-(lO) [VT-RU-81, 82; FS-RU-20-24]. Department of Anthropology, Report 47. Universityof Vermont, Burlington.

Yesner, D. N. Hamilton and R. Doyle, Jr.1983 "Landlocked" Salmon and Early HoloceneLacustrine Adaptation in Southeastern Maine. NorthAmerican Archaeologist, Vol. 4(4):307-333.

LATE ARCHAIC PERIOD(6000-2800 B.P.)

Regional Context

The Late Archaic has been called "one of our most fasci-nating and most confusing periods" in prehistory. It is "aperiod in which prehistory takes on many local expres-sions, and developmental themes interact as a series ofconfusing cross currents" (Snow 1980: 187,216). In broad-est detail, four major traditions are recognized in theregion of New England-New York-southern Quebec duringthe Late Archaic period: the Laurentian tradition; theNarrow Point tradition; the Susquehanna tradition; and theMaritime Archaic tradition. Although there is some tem-poral overlap during periods of transition, these tradi-tions are chronologically separated in time. Each tradi-tion persists for roughly a thousand years, betweenapproximately 6000 and 2800 B.P. The first three areclearly represented in the archaeological record ofVermont, while the Maritime Archaic tradition wasfocused along the northern Atlantic coast from Maine toLabrador. Based on often subtle changes in artifact assem-blages over time and in different geographical areas, tradi-tions are frequently subdivided into phases.

Because archaeologists have dealt more often with arti-facts from individual sites or from small numbers of sitesand not with whole cultural systems, these traditions andphases are defined almost exclusively on the basis of per-ceived differences in technologies and artifact assem-blages. There is an often implied assumption that such dif-ferences reflect culturally distinct groups of people, butthere are strong arguments against such a simplistic inter-pretation.


Laurentian Tradition

Regional Context

The Laurentian tradition, dated between 6000 and 4400B.P., is a complex of artifact traits in chipped stone,ground stone and bone which first appeared among popu-lations centered in the St. Lawrence Valley, includingwestern Vermont. Particularly diagnostic are Otter Creek,Brewerton and Vosburg projectile points and ground slateblades and ulus. It has been proposedthat the use of groundslate tools by pepole in Vermont derived from contactswith Maritime Archaic peoples who lived along the northAtlantic coast. and even that the presence of such tools inVermont represents an actual intrusion of MaritimeArchaic peoples into the Lake Champlain drainage(Haviland and Power 1981). Given the occurrence ofground slate points and groundstone celts and gouges at theSharrow site in interior Maine as early as 8300 B.P.,before the Maritime Archaic cultures developed, the migra-tion hypothesis of coastal peoples from eastern Canadarequires a reevaluation (Petersen 1989: 166-179).

Within the New York-southern Ontario-southernQuebec-northern New England region, three sub-region-al variants of the Laurentian tradition have been identi-fied. These three phases, called the Vergennes, Brewertonand Vosburg, overlap in time and space. The Vergennesphase is the oldest and is believed to be the base fromwhich the Brewerton and Vosburg phases developed.The Brewerton phase developed and remained focused incentral New York and in the upper St. Lawrence valleyaround Montreal. The Vosburg phase developed in theupper Hudson valley of eastern New York where itapparently succeeded the Vergennes phase by about 4800B.P. (Funk 1976:240; Ritchie 1969a:83).

Technology

Three sites in the Champlain Basin, Otter Creek No. 2and Kl (Ketchem Island) located along the middle reachesof Otter Creek, and the Bridge Site, located near themouth of the Little Ausable River on the west side of LakeChamplain, provide the best representation of Vergennesphase materials anywhere in the Northeast, and in a veryreal sense, these sites are the type sites for this phase(Figure 3). All sites contained a similar broad array ofhunting, fishing, woodworking and other processing tools(Ritchie 1979:8-17; Ritchie 1969a:86; Ritchie 1968:3).

From a roughly 130 m2 sample from Otter Creek No.2,

57


Figure 3. Late Archaic period sites cited in the text.

1. Highgate Convener Station2. Isle La Motte Cemetery3. IT-GI-184. Bridge5. Colchester Point6. Winooski7. McNeil Borrow8. Ewing9. Kl

10. Sumner's Falls11. Skitchewaug

0 10 2

1

0 30 40 !l0t==-J 1 1 1

MILES

0 10 ~o ;,~oB IKILOMETERS

58

dated to 5070 ± 210 B.P., flaked stone hunting and pro-cessing tools consist of Otter Creek and Vosburg projec-tile points, knives, drills, heavy chopping tools, and fivetypes of scrapers (Plate 3). Flaked stone tools were com-plimented by a bone spear point and by a number ofground slate projectile points or knives and crescent-shaped knives (ulus). Spears or lances were apparentlyused with spear throwers, as the assemblage also includesan atlatl weight. Fishing equipment includes plummets,gorges made from native copper or bone, and bone pointswhich have multiple barbs along one edge. Woodworkingtools consist of a variety of celts (ungrooved axes), adzes,gouges, and rod-shaped abrading stones which wereapparently used to resharpen the working edges or bits.Other items include hammers tones, mullers (possibly forgrinding plant foods), a graphite paint stone, bone awls,and deer antler tines which were probably used as flakersto make stone tools. Artifact assemblages indicate that byroughly 5100 B.P. communities living in the Champlainbasin utilized a diversified tool kit to perform a wide rangeof day-to-day activities including hunting, line, spear andprobably net fishing, the processing of plant foods andfibers, and various types of woodworking. The commonoccurrence of gouges suggests that dugout canoes, madeby hollowing out large tree trunks, were frequentlyemployed to take advantage of the food resources invarious river, pond and wetland systems, and possibly forlong- distance travel. There is also a notable concentrationof sites along navigable rivers, ponds and lakes.

Until recently, it was unclear whether a transition in


Vergennes phase artifact assemblages occurred in theChamplain Lowland of western Vermont or whetherVergennes phase assemblages persisted here for over athousand years. Funk (1988) had suggest ed a transition,and such an evolution seemed to be reflected at the Ewingsite on Shelburne Pond in Shelburne, Vermont.Excavations at VT-GI-18 on South Hero Island, whichproduced an artifact assemblage of Vosburg-like points,but which lacked Otter Creek and Brewerton points,ground slate and copper artifacts, indicate that such a tran-sition to a Vosburg phase did occur (Thomas, Kochan andDoherty 1992).

Based on a 63 m2 sample from two small residential areasin the southern portion of VT-GI-18, the artifact assem-blage consists of finished stone tools, including Vosburg-like projectile points, drills, and prepared scrapers; expedi-ency tools, such as modified flake scrapers and shavers,large flakes used for cutting or shaving, and pebble ham-merstones; a groundstone celt; portions of flaked toolsbroken during manufacture; and nearly 1,700 quartzite orchert flakes (Thomas, Kochan and Doherty 1992:52-78).


Little is known about Vergennes and Vosburg phase set-tlement and subsistence patterns, but long-term climaticchanges are likely to have brought about some culturalresponses. Beginning at roughly 5000 B.P., at the startof the Late Archaic period, and lasting- until approxi-mately 3600 B.P., the climate was characterized by warm

Plate 3. Vosburg-like projectilepoints in the Watson Collection fromChimney Point, Addison, Vermont.Photo by Stephen Loring.

59


summers and less intense winter temperature extremesthan during the preceding 1,500 years (Brackenridge et al.1988; Dincauze 1989). In Vermont, this warming trendallowed the gradual expansion of an oak-hemlock and thenan oak-hickory forest from the south. Nut trees and fruitshrubs became increasingly plentiful. Although peopleprobably concentrated on resources in the lower reaches ofthe major river valleys during the winter, the increasedfloral productivity throughout the region provided greaterseasonal flexibility in site location and exploitation of ani-mal and plant foods. As a result, settlements spread intoall physiographic regions of Vermont, encompassing all ofthe Champlain Lowland, as well as more favorable habi-tats in the Piedmont, Taconic and Green Mountains.

The functional characteristics of sites dating to thisperiod and their distributions are poorly understood. Onepossible circular house has been identified at the KIsite in the Otter Creek valley, but house forms are other-wise unknown. Although most sites may have been fairlysmall and occupied primarily by extended families, muchlike Otter Creek No.2, KI and VT-GI-18, it is clear thatpeople utilized a wide range of environments during thecourse of their seasonal rounds. Sites containing OtterCreek, Vosburg and Brewerton-like projectile points havebeen identified in all major watersheds in the ChamplainLowland and along stretches of the Hoosic andWalloomsac Rivers in southwestern Vermont. Vergennesand Vosburg phase sites are particularly prominent aroundthe extensive wetlands which border Otter Creek and itsmajor tributaries, Little Otter Creek and Lewis Creek.They are also common around lakes and ponds, such asLake St. Catherine, Lake Bomoseen, and Shelburne Pond,and have been recovered at sites on the shore of LakeChamplain south of the mouth of Otter Creek. There iscurrently little evidence of occupations in the upper endsof the major watersheds of western Vermont, along theConnecticut River or in the valleys of the ConnecticutRiver's major tributaries in eastern Vermont.

Although the distribution of sites suggests that people weretaking advantage of a wide range of food resources, foodremains have been recovered at very few sites dating tothis period. The Otter Creek No. 2 site located nearBrandon provides the best representative sample. Here,deer was the dominant species identified; all ages wererepresented. At least some of the kills were brought to thesite for butchering, so hunting within a short distance fromthe camp was apparently common. Bear, beaver, muskrat,turkey, great blue heron, and an occasional wood turtlewere also taken. Most identifiable bone fragments were

probably derived from large mammals, the long boneshaving been broken open for marrow extraction. No fishbones were recovered, but the presence of gorges, plum-mets, and barbed bone points clearly indicates that fishwere a regular part of the spring-fall diet. The processingand use of some plant foods are also suggested by thepresence of a muller (Ritchie 1979:17). Deer, bear, turtleand fish bones have also been identified from a smallhearth containing a Vosburg point at the Ewing site.Copper gorges used for fishing have been found here aswell (Petersen et al. 1985). Excavations at VT-GI-18produced only a few fragments of deer bone, but its loca-tion strongly suggests that the site's occupants alsoundertook fishing.

It has been suggested that people at this time followeda "central-based wandering community pattern," in whicha group spent a substantial portion of each year in one orseveral large base camps. The full community may havebeen composed of several hundred band members, theband itself being the major social and political unit withrights to use a particular territory. During the warmer sea-sons the band may have assembled where foodresources were particularly prolific; small task groups mayhave ranged out from the base camp to acquire specificresources. VT-GI-18 may represent such a small resi-dential camp occupied by a multi-family task group. TheEwing site on Shelburne Pond, which is characterized bythick midden deposits, may fall within the category of basecamp. Because other sites apparently contain a very nar-row range of artifacts, particularly projectile points, amore focused subsistence pattern based primarily onsmall group hunting may have been undertaken duringlate fall and winter.

Ideology

We may assume that people during the Late Archaic peri-od had a rich complex of spiritual beliefs. We are unlikelyto encounter much evidence of such beliefs in the surviv-ing archaeological record, except for limited informationabout how families may have treated the deceased.

Burials have been encountered at two Vergennes phasesites in Vermont. At Otter Creek No.2, Ritchie (1979:18-19) uncovered six very fragmentary skeletons which heattributed to this period of occupation. (Recent skeletalanalyses have raised questions about this interpretation,but the issue remains unresolved at this time.) The orien-tation of the bones of some individuals suggested toRitchie that these were secondary or bundle interments,

60

i.e., that after the bodies had decomposed elsewhere,most of the bones were collected, then brought to this sitefor burial. Definite grave goods were lacking, but a mid-dle aged man was clearly buried with a large dog. Reportsof a burial at the KI site are more sketchy. During the1950s, Tom Daniels uncovered what he took to be a struc-ture defined by a rough circle of post molds. Within thiscircle and under a covering of quartzite cobbles, Danielsfound a single decayed human skeleton, lightly sprinkledwith red ochre. Daniels also observed that the presumedhouse was overlain by a low earthen mound (Ritchie1969a:86). It is possible that the low mound underlain bya layer of cobbles and the ochre-covered skeleton are allparts of the same burial. Given the lack of adequate sitedocumentation, we will never know, but it is intriguingthat similar burial customs are known from areas on thecoast far to the northeast (Tuck and McGhee 1976; Tuck1977).

Narrow Point Tradition

Regional Context

The Narrow Point tradition, dated from roughly 4400 to3400 B.P., is represented by a number of complexes whichextend along the East Coast from North Carolina to Maineand as far north as the upper St. Lawrence River nearMontreal. From observed variations in artifact assem-blages in different geographical areas, a number of phas-es have also been introduced into the archaeological liter-ature. Based on data from the Sylvan Lake Rockshelterand other sites in the Hudson River watershed whichextends into southwestern Vermont, Funk postulates aSylvan Lake complex or phase, dated between approxi-mately 4400 and 3900 B.P. in the upper Hudson and LakeGeorge area. Between 3900 and 3400 B.P., artifact assem-blages at some sites seem to reflect a transition in whichNormanskill points become the dominant projectile pointtype, although Sylvan Stemmed and Sylvan Side-Notchedpoints are present. Funk proposes the term River phase forthese sites which he believes date from approximately3900 to 3350 B.P., thus overlapping with the Susquehannatradition on the coast (Funk 1976:247-251).

Technology

A variety of projectile points with relatively long and nar-row blades, generally weak shoulders, and straight,expanding and side- or corner-notched stems are the pri-mary diagnostic feature of the Narrow Point tradition. InVermont, well-documented assemblages of artifacts related

The Journal ofYermont Archaeology

solely to the Narrow Point tradition are virtually unknown.While this tradition is well represented in at least one dis-crete portion of the Ewing site on Shelburne Pond inShelburne, Vermont, no single component site has everbeen excavated (Figure 3). Based on extensive surface col-lections of artifacts from throughout western Vermont,however, this tradition is clearly represented in all majorwatersheds by a variety of projectile point forms typi-cal of northeastern New York and the upper St.Lawrence drainage (Ritchie 1961).

There are also a number of narrow stemmed andnotched points in Vermont collections which do not falleasily into any formal type (Squire 1977; Mills 1984).This may be explained in part by Funk's observationsat the deeply stratified Sylvan Lake Rockshelter in theHudson Valley. Funk concluded that all the narrowstemmed points in this stratum fell into a singe series ofintergrading varieties, that both stemmed and side-notched points were contemporaneous, and that theyhad been made and used by a single people. Rather thanattempting to define various point types, "the evidentunity in basic characteristics of the stemmed points seemedto require the postulation of one inclusive, synthetic group,Sylvan Stemmed, a stylistic continuum from which couldbe selected specimens bearing correspondences to typespreviously isolated in other parts of the Northeast;Lamoka, Bare Island, and the sizeable residue of"Lamoka-like" points" (Funk 1976:247-248). Narrowstemmed points and slightly broader side- notched pointsfound in Vermont are likely to reflect the same type of vari-ability as parts of single artifact assemblages, and mayormay not be chronologically distinct (Plate 4).

Identified sites of this period appear to have a more limit-ed range of artifact types than earlier Laurentian sites."On excavated components, projectile points outnumberother individual trait categories, sometimes predominatingover all other items combined" (Funk 1976:251-252). Inaddition to a variety of small projectile points, recoveredartifacts include thick bifacial knives, retouched flakeknives, drills, retouched flake side and end scrapers, pes-tles, possible choppers, netsinkers, pebble hammerstones,whetstones, anvilstones, graphite paint stones, and antlerflaking tools. Large, groundstone woodworking tools arerare. which suggests a shift from dugouts to bark canoes,although dugouts were never entirely replaced. A fewgroundstone tools have been found on River phase sitesand a few atlatl weights or bannerslones have also beenrecovered (Funk 1976:247-256).

61

Vermont Archaeulogy Comes of Age

Plate 4. Narrow Point traditionstemmed projectile points inthe Watson Collection fromChimney Point, Addison,Vermont. Photo by StephenLoring.

Whether sites dating between roughly 4400 and 3350B.P. in western Vermont contain similar or different arti-fact assemblages remains to be determined. It is possiblethat sites of this period in eastern Vermont may be moreclosely related to the contemporaneous Squibnocket com-plex or phase of southern New England (Ritchie 1969b).

In southern New England, Dincauze (1973, 1974) hasnoted a tendency for communities to use only immediate-ly available stone for tool manufacture, thus suggestingthat little regional exchange of raw materials or finishedproducts took place. At a regional level, we see a sub-stantial shift from southern New England and coastal NewYork where points are made almost entirely of quartz tothe Champlain Valley where contemporaneous points aremade predominantly from quartzite and chert. Furtherresearch may define even greater subregionalism in the useof raw materials among watersheds in Vermont.


During the period characterized by the Narrow Point tra-dition, a repetitive pattern in the general characteristicsand distributions of sites has been observed throughoutsouthern New England and New York. Sites, and there-fore community sizes, are generally not large, but they arenumerous and occur in a wide variety of local settings. Insouthern New England, Dincauze (1975:25) interprets alargely similar pattern as indicating a very diffuse subsis-tence adaptation, a relatively dense population, and acomplex pattern of seasonal movement.

Close parallels may exist in western and probably in east-

ern Vermont. Artifact assemblages from multicomponentsites occupied throughout the Late Archaic period areoften dominated by narrow stemmed points. For exple,118 projectile points were collected by Levi Pratt from 15sites located on Lake St. Catherine and along Wells Brookwhich connects the lake to the Mettawee River. Of thesepoints, four Otter Creek and seven Brewerton Side-Notched points represent Laurentian occupations, while73 narrow stemmed and side-notched varieties representa quantum jump in point frequency at Narrow Point tradi-tion occupations. Susquehanna tradition occupations arerepresented by 11 Snook Kill points. Not until the lateMiddle-Late Woodland periods a re points againencountered at sites in this area (Thomas and Bayreuther1980:14). A parallel jump in the number of narrowstemmed points versus earlier Laurentian and laterSusquehanna tradition points is evident at sites alongportions of the Walloomsac, Hoosic, Lemon Fair Riversand Dead Creek as well (Thomas, Campoli and Doherty1979:60,85; Thomas and Doherty 1980:62-63; Thomasand Doherty 1982:36-38).

As during the preceding period, all major stream valleyswhere food resources were particularly dense andlordiversified were intensively occupied, but new patternsemerged and old patterns intensified. Once marginalupland areas were now exploited on a more persistentbasis. Sites containing small-stemmed points appear forthe first time in the upper White River valley, along thePassumpsic River and on its major tributary streams, on theshore of Lake Memphremagog, and in the upper BlackRiver watershed on Amherst Lake (Doherty, Kochan andThomas 1990:7; Thomas, Doherty and Warren 1982:37;

62

Plate 5. Cache blades fromColchester Point in the R.S.Peabody Foundation Collection.Photo by Bruce Bourque.


Thomas et al. 1981:65; Thomas and Torrence 1987:11-12). Sites also appear in new settings, including uplandridges located well above the valley in the Winooski andpossibly the Wells River watersheds (Thomas 1989:11;Thomas and Torrence 1987:12). In the lowlands, sec-ondary and even minor stream valleys contain numeroussites dating to this period (Thomas, Dillon and Doherty1985). In short, this period is characterized by the mostvaried and complex settlement pattern during the entireArchaic and most of the subsequent Woodland period.

Subsistence data from Narrow Point tradition sites inVermont are virtually unknown. Presumably, many ofthe basic Vergennes patterns persisted, since componentsof all three traditions are commonly encountered at thesame sites. The only dated food remains for this periodwere recovered at a depth of 2.75 m (9 ft) from a hearth atthe McNeil Borrow site on the lower Winooski Riverfloodplain. Well over 1,000 pieces of burned mammalbone, including deer, were found in this feature dated to3670 ± 150 B.P. (Thomas and Robinson 1981). Deer,bear, turtle and fish bones were also apparently associatedwith a feature containing small-stemmed points at a site onShelburne Pond (Petersen et al. 1985).

Ideology

Only one possible Narrow Point tradition burial has beenidentified in Vermont, and this is inferred only by thepresence of a unique cache of about 25 blades found in the

1920s in a small pit on Colchester Point (plate 5). Theselarge blades, characterized by long straight stems andflared bases, are rare in Vermont. However, points ofvery similar form appear almost exclusively in mortuarycontexts in Maine where they date to approximately 3950B.P. and are attributed to the Moorehead phase. Notably,some of the Bradley points from Colchester Point may bemade from Kineo rhyolite from Maine and various Bradleypoints found in Maine are clearly made from Cheshirequartzite from Vermont (Bourque 1971:91-93).

Susquehanna Tradition

Regional Context

The Susquehanna tradition, dating from roughly 3800 to2800 B.P., is represented in the archaeological record by anumber of related cultural complexes which have alsobeen termed the Broadspear tradition, or the''Transitional'' or ''Terminal Archaic" stages (Funk 1976;Ritchie 1969a). It now seems clear that the Susquehannatradition originated in the Southeastern Piedmont as earlyas 4500 B.P. and diffused northward as far as the coast ofcentral Maine and into the upper St. Lawrence valley ofsouthern Quebec by approximately 3600 B.P. (Bourque1971; Clermont and Chapdelaine 1982:35-41; Coe 1964).The ultimate demise of the Susquehanna tradition, or ofthe intrusive peoples who some believe actually broughtthis artifact tradition with them as they moved into north-

63


em New England, occurred about 2800 B.P. (Bourque1975; Dincauze 1968, 1975; Sanger 1975). Little isknown, however, about the circumstances leading to thedisappearance of this cultural complex or the nature ofchanges giving rise to the subsequent Woodland way oflife.

Technology

Based on radiocarbon dates, the Susquehanna tradition inNew England persisted for approximately 900 years,between roughly 3650 and 2800 B.P. Projectile pointsare again the most diagnostic items in the artifact assem-blages (Plate 6). A variety of generally large, broad-blad-ed points with straight to slightly contracting stems aretypical and are known variously as Atlantic, Snook Kill,Batten Kin, Perkiomen, Genesee and Susquehanna Broad.Snook Kill points appear earliest, while a smaller point,known as Orient Fishtail, is characteristic of the end of thetradition, ca. 3050-2800 B.P. (Funk 1976:252-·263).

Beginning about 3250 B.P., carved soapstone or steatitepots occur at sites in association with Susquehanna Broadpoints, and later with Orient projectile points. By rough-ly 2850 B.P., the first pottery was being made in theNortheast, a type corrunonly referred to as Vinette 1(Belcher 1989). In addition to these new types of vessels,habitation sites of this period have produced a range of arti-facts including atlatl weights, end and side scrapers,drills, bifacial knives, flake knives, utilized flakes, ham-

mer-anvilstones, abrading stones, whetstones, pestles andnotched netsinkers. Although their presence has not beendocumented in dated contexts in the Hudson Valley andin Vermont, grooved axes, celts, adzes, gouges and paint-stones are associated with Orient phase sites on LongIsland (Funk 1976:261-267; Ritchie 1969a:175-177).

In Vermont, similar broad-bladed and Orient Fishtail-like points of the Susquehanna tradition have been recov-ered during surveys of plowed fields and at multi compo-nent sites throughout western Vermont; burials areunknown (Squire 1977; Mills 1984). Except for one focalarea at the Highgate Converter Station site on theMissisquoi River, however, single component sites datingto this period have not been excavated and artifactassemblages are poorly defined (Figure 3). In Highgate, a22 m2 block sample from what appears to be a small hunt-ing camp exposed a shallow hearth dated to approximate-ly 3200 B.P. Nine flake tools utilized for cutting.shaving and scraping. several tool manufacturing areascontaining over 750 chert flakes, the tip of a brokenprojectile point and a chert preform were found inclose association. Unfortunately, no complete projectilepoints were recovered which could be used as dated typesamples (Thomas and Dillon 1985:59-83).

Plate 6. Broad bladed points inthe Watson Collection fromChimney Point, Addison,Vermont. Photo by StephenLoring.

64


Susquehanna tradition sites appear to be less numerousthan the earlier Laurentian and Narrow Point traditionsites and generally contain lower artifact densities. Rarelyare more than a few points recovered at an individual sitein Vermont. Susquehanna tradition sites are also morecommon in southwestern Vermont in the upper Hudsondrainage than they are in the north. RecordedSusquehanna tradition sites relative to all sites of knownage in the state site files range from a high of 6.3% inBennington County to a low of only 1.3% in FranklinCounty (Thomas and Dillon 1985:82-83).

Contemporaneous sites in the Connecticut River valley arerare, but confirmation of their presence was recoveredfrom a high alluvial terrace at Sumner's Falls on theConnecticut River in Hartland. In the deepest stratumexcavated, artifacts consisted of four Orient-like projectilepoints and a small cooking pit from which the charcoalwas dated to 2750 ± 80 B.P. Additional evidence ofSusquehanna tradition sites comes from surface finds oftools and fragments of steatite pots from the area ofSkitchewaug along the Connecticut River in Springfield(Haviland and Power 1981:81-82; Heckenberger andPetersen 1988:86). Based on current evidence, little canbe said of the social system characteristic of theSusquehanna tradition populations in Vermont. Thecontinuation of a central-based wandering system of thepreceding period seems likely. The household unit proba-bly continued to be the single nuclear or small extendedfamily, and the settlement units probably continued to besmall but numerous (Snow 1980:249). However, someshift in the distribution of sites is evident.

At about 3000 B.P., the climate began a reversion tonoticeably cooler and moister conditions. The percent-ages of deciduous trees such as oak, beech and mapledropped, while hemlock, pine and birch increased in gen-eral proportion. Perhaps the density and diversity of gamespecies was also somewhat constricted. The ubiquitousresources of the previous 1,500 years were, therefore,becoming less available. Although riverine, lakeshore andwetland sites in the major valleys of the ChamplainLowland, Vermont Valley, upper Hudson Lowland andConnecticut Valley continued to be occupied, there is anoticeable drop in the number of sites and density of arti-facts at upland sites throughout Vermont as theSusquehanna tradition period wound to a close.Seemingly, throughout Vermont, the upland regions werenearly abandoned by people after roughly 2800 B.P.


Ideology

From southern New England to the central Maine coast, theSusquehanna tradition is associated with an elaboratepattern of mortuary ceremonialism which emphasizedthe practice of cremation, the ritual use of red ochre, andthe often lavish inclusion of grave goods, including atlatlweights, projectile points of very high quality material,and woodworking tools. Evidence for elaborate burialceremonialism is not reflected at Susquehanna traditionsites in Vermont.

Glacial Kame Burials

At the very end of the Late Archaic period, but unrelatedto the Susquehanna tradition, burial ceremonialism is rep-resented by a unique cemetery located on a gravel ridge onIsle La Motte (Figure 3). Two in situ burials were foundin shallow pits containing burned and unburned bonestained with red ochre. Six and possibly 12 additionalgraves may have existed before gravel operationsdestroyed the site. Three distinctive sandal sole gorgetsmade from marine shell were found at Isle La Motte.These are diagnostic of the Glacial Kame burial com-plex which is focused in the south central Great Lakesregion of Michigan, Ohio, Indiana and the Province ofOntario. Radiocarbon dates of 2930 ± 80 B.P. from Isle LaMotte and 2975 ± 75 and 2920 ± 170 B.P. for Glacial Kameburials at the Hind site far to the west are nearly identical.In addition to the sandal sole gorgets, other exotic goodsdrawn from hundreds of mile away were added as graveofferings, including copper adzes, rolled copper beadsstrung on a cord made from either milkweed or Indianhemp fibers, lumps of unworked galena, discoidal shellbeads, a circular three-hole gorget of marine shell, a rec-tangular gorget of walrus tusk, fragments of a possibleleather shroud, and the tip and midsection of a Meadowoodprojectile point (Haviland and Power 1981:83-85; Power1989; Ritchie 1969a:132-135).

The Isle La Motte cemetery marks the far easternfringe of the known distribution of Glacial Kame burialpractices. It is an enigma of sorts, as a number of theburial practices are similar to those represented at earlierMoorehead phase (Red paint) cemeteries of Maine andin only slightly later Meadowood-Middlesex phase buri-als in the Champlain Lowland of Vermont. Who thesepeople were and how this cemetery relates to the devel-opment of cultural systems in the Champlain Valley willprobably remain a mystery, because it seems unlikely thatother Glacial Kame cemeteries will ever be discovered.

65


Conclusions

A great deal of variability probably exists amongLaurentian, Narrow Point and Susquehanna tradition sitesof the Late Archaic period; change within this 3,300 yearperiod seems likely. Factors such as seasonality of siteoccupation, the types of re sources which could be exploit-ed from a particular setting, and the specific types ofstrategies which people used to hunt, fish, gather, consumeand possibly store various food items would all affectthe types of archaeological remains which are likely tohave been left at specific sites.

Obviously, much remains to be learned about how vari-ous types of Late Archaic sites were integrated into largersettlement systems and how such systems may havechanged through time. Sites in a number of environmentswill be difficult to locate. Excavations as the Winooskiand McNeil Borrow sites and an extensive survey con-ducted along the levee at the lower end of the MissisquoiRiver, for example, confirm that Late Archaic sites insuch settings are likely to be deeply buried. At theWinooski site, occupations dating to 3450 ± 175 B.P. and4260 ± 235 B.P. were identified at depths of 1.7 m (5.6 ft)and 2.3 m (7.5 ft), respectively. At the McNeil Borrowsite, a hearth dated to 3670 ± 150 B.P. was found at adepth of 2.75 m (9 ft). Based on the presence of deeplystratified but relatively young sites on the floodplainsalong the lower Missisquoi and Connecticut Rivers, manyLate Archaic period sites here may be buried below 2 mor more of flood deposits. One deep stratigraphic unitdated to ca. 3050. B.P. at the Skitchewaug site on theConnecticut River supports this contention (Heckenbergerand Petersen 1988; Petersen and Power 1983; Thomasand Robinson 1979, 1981). Other sites may have beensubmerged beneath Lake Champlain. Detailed and exten-sive research at sites in all physiographic areas will berequired before more than a general outline of tech-nological changes and settlement and subsistence patternscan be reconstructed.

Late Archaic Period: References

Belcher, W. R.1989 Prehistoric Fish Exploitation in East Penobscot Bay,Maine: The Knox Site and Sea-Level Rise. Archaeologyof Eastern North America, Vol. 17:175-190.

Bourque, B. J.1971 Prehistory of the Central Maine Coast. Ph.D. dis-sertation, Department of Anthropology, Harvard

University.

1975 Comments on the Late Archaic Populations ofCentral Maine: The View from Turner Farm. ArcticAnthropology, Vol. 12 (2):35-45.

Brakenridge, G. R., P. A. Thomas, L. E. Conkey and J. C.Schiferle1988 Fluvial Sedimentation in Response to PostglacialUplift and Environmental Change. Quaternary Research,Vol. 30:190-203.

Clermont, N. and C. Chapdelaine1982 Pointe-du-Buisson 4: Quarante Siecles d'ArchivesOubliees. Recherches Amerindiennes au Quebec,Montreal.

Coe, J. L.1964 The Formative Cultures of the Carolina Piedmont.Transactions of the American Philosophical Society, Vol.54(5).

Dincauze, D.1968 Cremation Cemeteries in Eastern Massachusetts.Papers of the Peabody Museum of Archaeology andEthnology, Vol. 59(1). Harvard University.

1973 Prehistoric Occupation of the Charles RiverEstuary. Archaeological Society of Connecticut Bulletin,Vol. 38:25-39.

1974 An Introduction to Archaeology in the GreaterBoston Area. Archaeology of Eastern North America, Vol.2:39-66.

1975 The Late Archaic Period in Southern NewEngland. Arctic Anthropology, Vol. 12 (2):23-24.

1989 Geoarchaeology in New England: An EarlyHolocene Heat Spell? The Review of Archaeology, Vol.10(2):1-4.

Doherty, P., G. Kochan and P. A. Thomas1990 Archaeological Reconnaissance Survey forLyndon RS 0259(9), Lyndon, Vermont. ConsultingArchaeology Program, Report 99. University ofVermont, Burlington.

Funk, R. E.1976 Recent Contributions to Hudson Valley Prehistory.New York State Museum and Science Service Memoir 22.

66

1988 The Laurentian Concept: A Review. Archaeology ofEastern North America, VoL 16:1-42.

Haviland, W. A and M. Power1981 The Original Vermonters: Native Inhabitants Pastand Present. Hanover: University Press of New England.

Heckenberger, M. J. and J. B. Petersen1988 Archaeological Investigations at the SkitchewaugSite: A Multicomponent Stratified Site in Springfield,Windsor County, Vermont. Report prepared for theVermont Division for Historic Preservation,Archaeological Research Center, University of Maine atFarmington, Farmington, Maine.

Mills, P. R.1984 A Refinement of the Typologies of VermontProjectile Points and Other Chipped Stone Tools in theChamplain Valley. BA thesis, Department ofAnthropology, University of Vermont.

Petersen, 1. B.1989 Archaeological Testing at the Sharrow Site: ADeeply Stratified Early to Late Holocene CulturalSequence in Central Maine. Archaeological ResearchCenter, University of Maine at Farmington, Farmington,Maine.

Petersen, 1. B. and M. W. Power1983 The Winooski Site and the Middle Woodland Periodin the Northeast. Report submitted to InteragencyArchaeological Services, Mid-Atlantic Region, NationalPark Service.

Petersen, J. B., J. A Wolford, N. D. Hamilton, L. LaBarand M. Heckenberger1985 Archaeological Investigations in the Shelburne PondLocality, Chittenden County, Vermont. Annals of theCarnegie Museum, Vol. 54(3):23-75.

Power, M.1989 The Isle La Motte Site: A Revisionist's View ofGlacial Kame. Paper presented at a Symposium onRethinking the Terminal Archaic- Woodland Period in theNortheast, Annual Meeting of the Society for AmericanArchaeology, April 6, Atlanta.

Ritchie, W. A.1961 A Typology and Nomenclature for New YorkProjectile Points. New York State Museum and ScienceService Bulletin 384.


1968 The KI Site, the Vergennes Phase and the LaurentianTradition. New York State Archaeological AssociationBulletin, Vol. 42:1-5.

1969a The Archaeology of New York State [revised edi-tion]. Garden City: The Natural History Press.

1969b The Archaeology of Martha's Vineyard. GardenCity: Natural History Press.

1979 The Otter Creek No.2 Site in Rutland County,Vermont. The Bulletin and Journal of the Archaeology ofNew York State, Vol. 76:1-21.

Sanger, D.1975 Cow Point: An Archaic Cemetery in NewBrunswick. National Museum of Man, Mercury Series 12,Ottawa.

Snow, D.1980 The Archaeology of New England. New York:Academic Press.

Squire, M.1977 Classification of Projectile Points from theChamplain Valley in Vermont. Unpublished MA thesis,Department of Anthropology, State University of NewYork, Buffalo.

Thomas, P. A1989 Phase 1A Archaeological Reconnaissance Survey:Waterbury-Duxbury-Moretown BRF 013-4(10),Washington County, Vermont. Consulting ArchaeologyProgram, Report 101. University of Vermont, Burlington.

Thomas, P. A and W. A. Bayreuther1980 Wells RS 0145(7): Phase I ArchaeologicalAssessment. Department of Anthropology, Report 25.University of Vermont, Burlington.

Thomas, P. A, G. Campoli and P. Doherty1979 An Initial Study of the Prehistoric and HistoricArchaeological Sensitivity of the Bennington Area.Department of Anthropology, Report 22. University ofVermont. Burlington.

Thomas, P. A, G. Campoli, P. Doherty and J. B. Petersen1981 Phase I Archaeological Assessment for theMissisquoi Hydro-electric Project: Sheldon Springs,North Sheldon, Enosburg Falls, North Troy and Jay

67


Branch Elements. Department of Anthropology, Report34. University of Vermont, Burlington.

Thomas, P. A. and R. S. Dillon1985 The Highgate Converter Station Site. Department ofAnthropology, Report 55. University of Vermont,Burlington.

Thomas, P. A., R. S. Dillon and P. Doherty1985 Archaeological Reconnaissance Survey for theChittenden County Circumferential Highway. VolumeII: Site Description and Technical Information.Department of Anthropology, Report 52. University ofVermont, Burlington.

Thomas, P. A. and P. Doherty1980 Archaeological Resource Management Study:Lower Otter and Dead Creek Watershed. Department ofAnthropology, Report 26. University of Vermont,Burlington.

1982 Archaeological Resource Management Study,Lemon Fair River Watershed. Department ofAnthropology, Report 40. University of Vermont,Burlington.

Thomas, P. A., P; Doherty and E. J. Warren1982 Cultural Resource Management Study, BallMountain Lake, Springfield and Weathersfield,Vermont. Department of Anthropology, Report 38.University of Vermont, Burlington.

Thomas, P. A., G. P. Kochan and P. Doherty1992 Windows to the Past: Archaeological Excavationsat the Grand Isle Fish Hatchery, Grand Isle, Vermont.Archaeology Program, Report 115. University ofVermont, Burlington.

Thomas, P. A. and B. S. Robinson1979 Missisquoi National Wildlife Refuge: A CulturalResource Survey. Department of Anthropology, Report10. University of Vermont, Burlington.

1981 Joseph C. McNeil Generating Station: BorrowArea. Archaeological Evaluation [VT -CH-146].Department of Anthropology, Report 31. University ofVermont.

Thomas, P. A. and C. McP. Torrence1987 Archaeological Reconnaissance Survey forGroton-Ryegate F-020-1(30), Groton and Ryegate,

Vermont. Department of Anthropology, Report 79.University of Vermont, Burlington.

Tuck, J. A.1977 A Look at Laurentian. In Current Perspectivesin Northeastern Archaeology-Essays in Honor ofWilliam A. Ritchie, edited by R. E. Funk and C. F. Hayes,pp.31-4O. Researches and Transactions of the New YorkArchaeological Association, Vol. 17(1).

Tuck, J. A. and R. J. McGhee1976 An Archaic Indian Burial Mound in Labrador.Scientific American, Vol. 235 (5):122-129.

EARLY WOODLAND PERIOD(2800 - 1850 B.P.)

Regional Context

A number of salient features have been recognized for theEarly Woodland period. Beginning around 2950 B.P. in theMid-Atlantic and about 2800 B.P. in the New Englandarea, the first significant use of pottery can be document-ed (Belcher 1989; Petersen and Hamilton 1984). Long-distance exchange networks were well established on ascale rarely if ever exceeded before or after in the region(Ritchie 1969). Unlike areas in the upper Mississippi andOhio valleys where domesticated crops and food produc-tion began to play an increasingly important role between2500 and 1750 B.P. (Smith 1989), there appear to havebeen no significant shifts in annual subsistence patternsin the Northeast where the food economy continued to bebased on hunting, fishing and gathering during the EarlyWoodland period. And, unlike preceding Late Archaicsettlement patterns in which sites are present in a wide vari-ety of physiographic settings, general use of the regionallandscape seems to have been more restricted (Funk1976:278; Ritchie 1969:179-181, 190).

It is ironic that while burial sites dating to most periods inprehistory are uncommon, the Early Woodland period isbest known and almost totally defined on the basis of datarecovered from complex cemeteries. One of the major fea-tures of the Early Woodland period is the continuationand elaboration of a well-developed mortuary ceremonial-ism which had strong roots in the cultures of the preced-ing Late Archaic period.

William Ritchie identified two Early Woodland phases in

68

New York: the Meadowood phase and the Middlesexphase. In areas outside of New York. attempts to adoptRitchie's terminology have caused considerable confusionin the archaeological literature, largely because the con-cepts are so ill defined and because more recent radiocar-bon dates do not support a chronological sequence fromMeadowood to Middlesex. Based on current evidence, itseems reasonable to use the term Middlesex to refer not toa wider cultural system, but exclusively to a mortuary com-plex found throughout the far Northeast, including east-central and eastern New York, New England, Quebec andthe Maritime Provinces of eastern Canada. Based on 24radiocarbon dates ranging from 2835 to about 2050 B.P.(and possibly later) from the Boucher cemetery (Vermont)and six dates ranging from 2950 to 2330 B.P. atAugustine Mound (New Brunswick), the Middlesex mor-tuary complex apparently characterized the entire EarlyWoodland period in the far Northeast and must have beencontemporaneous with the Meadowood phase in NewYork (Heckenberger, Petersen and Basa 1990). Strongsimilarities also exist between many aspects of theMiddlesex mortuary complex and slightly earlier GlacialKame burials encountered on Isle La Motte.

Technology

Early Woodland period residential sites are likely to con-tain stemmed or side-notched projectile points which areusually referred to as Adena and Meadowood, respective-ly. Thin, side-notched Meadowood points occur in west-ern Vermont, as well as at a few sites in the Connecticutwatershed, but only one or two specimens are generallyrecovered from anyone site or found in individual arti-fact collections. In western Vermont, the occurrence oflobate-stemmed Adena points in artifact collections fromall watersheds is uncommon; Adena points are rare inthe Connecticut valley of eastern Vermont. A third type ofprojectile point, called a Lagoon point, has been recoveredin western Vermont. This style is far more typicallyencountered at coastal sites in southern New England.

In the Champlain valley, portions of two small residen-tial sites have been excavated: VT-FR-161 and VT-CH-234 (Thomas and Dillon 1985; Petersen,Heckenberger and Thomas 1988). Two Vermont siteslocated on the Connecticut River floodplain are alsoknown to contain Early Woodland occupations:Skitchewaug in Springfield and the Canaan Bridge site inCanaan (Figure 4). A radiocarbon date of 2670 ± 70B.P. from deep in the stratigraphic sequence undoubtedlydefines an Early Woodland occupation at the


Skitchewaug site. Related materials include severalMeadowood projectile points, a broken Meadowoodcache blade, a possible Meadowood scraper, and a heav-ily reworked lobate-stemmed Adena-like projectile point,along with Vinette I ceramics (Heckenberger andPetersen 1988:86).

Excavations at the Canaan Bridge site farther upstream onthe Connecticut River provide the most detailed informa-tion yet on an Early Woodland habitation site in Vermont(Bolian and Gengras 1991). Ten cultural features, most-ly shallow hearths characterized by fire reddened andcharcoal infused soil and fire shattered rock; appear tobe related to several occupations. Ten associated radiocar-bon dates range from 2320 to 2860 B.P. Given the stan-dard deviations related to these radiocarbon dates, most ofthe occupational events could have occurred betweenabout 2500 and 2650 B.P.

The artifact assemblage from a 59 m2 area may be fair-ly typical for small residential camps where hunting, fish-ing and gathering activities appear to be represented.Overall, the artifact density is low. Fragments fromthree ceramic vessels were recovered; all have classicVinette I characteristics. Lithic artifacts include: onereworked Meadowood projectile point; part of a possiblechert cache blade; eight chert and two rhyolite scrapers;16 chert flakes utilized for either scraping or cutting; onegraver; two hand grinding stones; one grinding slab; onepolishing stone; one pebble harnmerstone; one haftedhammer; and a gouge-like pebble with one large flakeremoved. The only decorative item recovered is part of ahighly polished, ground stone pendant.

The vast majority of lithic debitage consists of small, softhammer or pressure flakes related to tool refurbishmentor edge maintenance. Nearly all of the tools and 96.4% ofthe flakes derive from some type of chert, the most com-mon being red, red with black bands or red mottled withgreen. Bolian and Gengras (1991) suggest two possiblesources, the Jim Pond Formation of west central Maineand Quebec, and the Munsungun Lake Formation innorthern Maine. Other cherts could derive from westernVermont or eastern New York. The remaining flakes arevolcanics, probably rhyolites derived from various sourceareas to the south and east. This exclusive use of non-localstone by the Early Woodland period occupants of theCanaan Bridge site reinforces the notion that both short andlong distance exchange networks which insured the acqui-sition of elaborate mortuary goods at this time also oper-ated to insure the distribution of more mundane materials

69


Figure 4. Early Woodland sites cited in the text.

1. Boucher2. VF-FR-163. VF-FR-484. Swanton (Frink Farm)5. VF-FR-1616. VT-CH-2347. East Creek8. Bennett9. Canaan Bridge

10. Woodstock11. Sldtchewaug

MILES

Ir~oKILOMETERS

70

for everyday use.

The Early Woodland period in Vermont and throughoutNew England is best known from mortuary sites.Middlesex burials often contain rather spectacular arti-facts, many of which could have been acquired onlythrough interregional exchange. Typical stone artifactsinclude blocked-end tubes of Ohio fireclay or steatite,lobate-stemmed Adena and side-notched Meadowoodprojectile points, bifacial scrapers, gravers, trianguloidleaf-shaped Meadowood cache blades, gorgets, pendants,paintstones, boatstones, birdstones, discoidal stones("chunkey stones"), bar amulets, plummets, celts, whet-stones, sinewstones, and nodules of iron and ironpyrite. While the non-local nature of the raw materialsencountered is often noted, including grey and brown mot-tled Onondaga chert (New York), Mistassini quartzite(central Quebec), and Harrison County (Indiana) and FlintRidge (Ohio) cherts, the use of local quartzite and chertmaterials should not be overlooked.

Vinette I vessels with smoothed-over interior and exteri-or surfaces and an incised triangular design about mid-body were recovered at both the Boucher and East Creekcemeteries. Bone artifacts include needles, punches,awls, fishhooks, antler tine flakers and hafted beaverincisors. Tubular and disc shell beads derived fromOlivella, Marginella, quahog and whelk, all marinespecies, were frequently added as grave goods, while shellpendants were less common. Copper items includeawls. celts and chisels. but a common characteristic ofmany Middlesex burials is the extensive use of rolledand tubular copper beads. some of which had been strungon twisted vegetable fiber cordage or on spun andunspun thongs (Haviland and Power 1981; Heckenberger,Petersen and Basa 1990; Loring 1985).

The inclusion of copper beads and the subsequent pro-duction of copper salts through oxidation created veryfavorable conditions for the preservation of bone andorganic artifacts in a number of graves. "In fact, the extantassemblage from [the Boucher] cemetery constitutes oneof the largest collections of perishable artifacts of thisantiquity known from anywhere in eastern NorthAmerica" (Heckenberger, Petersen and Basa 1990: 180).Hide artifacts include knotted thongs, "medicine" bagscontaining snake and animal skeletons, and a tailored hideshirt. Fragments of single, two, three and four ply cordagemade from milkweed, Indian hemp, basswood bast fIberand nettle have survived. Shrouds or cowls, mats andbags were made from twined textiles of plant fibers and


animal hair (Heckenberger, Petersen and Basa 1990).This assemblage of perishable items from the Boucher sitehas, for the first time, provided solid evidence of the com-plicated manufacturing techniques involved and a rareglimpse at the broad range of sophisticated hide and textileitems used by people in the Champlain basin some 3,000years ago.


At about 2950 B.P., as the Early Woodland period began,pollen cores recovered from bogs throughout Vermontand adjacent areas indicate that a climatic shift occurred.The percentages of deciduous trees such as oak, beechand maple dropped. Hemlock, pine and birch increased ingeneral proportion. The return of spruce and fir at higherelevations and on north-facing slopes indicates that a cool-er climate prevailed. Deciduous trees such as oak becamerestricted to lower elevations, particularly in the valleybottoms and on south and west- facing hillsides. Nut bear-ing species of trees became more isolated.

Although archaeological data related to Early Woodlandperiod sites in Vermont are very limited, a pond-lake-riverine orientation of the few known residential sites isapparent. There is certainly evidence that groups duringthe Early Woodland period exploited the lower reachesof a number of major river drainages, including theMissisquoi, Lamoille, Winooski, Otter Creek, Lemon Fairand East Creek (Petersen, Heckenberger and Thomas1988; Ritchie 1970:lO; Thomas and Bayreuther 1979;Thomas and Dillon 1985; Thomas, Doherty and Dillon1983; Thomas et al. 1981). Residential camps are alsopresent on the Connecticut River floodplain fromSpringfield to as far north as Canaan just south of theCanadian border (Bolian and Gengras 1991; Heckenbergerand Petersen 1988). With the possible exception of a poor-ly excavated, multicomponent site on East Creek, substan-tial habitation sites which might represent large basecamps have not been recognized anywhere. On the basisof current information, upland regions throughoutVermont were nearly abandoned by people after about2950 B.P., with groups tending to remain in the relativelymore prolific lowlands. There may even have been a con-current drop in the region's human population.

Site locations provide at least one clue to resourceexploitation at this time. Such settings include the shoreof Lake Champlain, floodplains along major rivers, bluffsabove major rivers with adjacent tributaries, alluvial ter-races adjacent to major falls, lowland ponds and secondary

71


lowland streams. Such site locations tend to reinforce thehypothesis that subsistence patterns involved the integra-tion of hunting, fishing and gathering activities.

Stable isotope analyses of carbon and nitrogen from humanbone indicate a strong emphasis on hunting and fishing,with less emphasis on gathered plant foods which weremore characteristically eaten by people during the earlierLate Archaic and later Middle-Late Woodland periods(Krigbaum 1989). At the site specific level, data derivedfrom VT-CH-234 in Essex suggest that some sites wereutilized as small residential camps by perhaps 5-15 indi-viduals pursuing subsistence practices based on foraging.Within Area 5, a small roasting pit was found to contain1,987 pieces of calcined bone, including cranial and longbone fragments, and phalanges from one or more deer;and an ulna, fibula and carpals of a beaver. Deer andprobably snapping turtle carapace recovered in Area 4may also be related to Early Woodland period occupa-tions (Petersen, Heckenberger and Thomas 1988: 129-134). Hunting, particularly for deer or bear, also appears tohave been the dominant subsistence activity at VT-FR-161 in Highgate (Thomas and Dillon 1985). In addition,some of the faunal remains included in burials at theBoucher cemetery on the lower Missisquoi River wereundoubtedly foods placed in the graves. Animals repre-sented include moose, deer, bear, raccoon, beaver andwild turkey (Heckenberger, Petersen and Basa 1990).

In eastern Vermont, a mixed food economy is again indi-cated at what is presumed to have been multiple fall occu-pations at the Canaan Bridge site on the upper ConnecticutRiver floodplain. Beaver, deer, moose and unidentifiedbird bone were recovered from several Early Woodlandhearths. Grinding stones, as well as burned butternut orwalnut shells, strongly suggest the use of a variety ofplant foods (Bolian and Gengras 1991).

A great deal of variability undoubtedly existed in thefocus of subsistence activities at Early Woodland residen-tial sites. The season of site occupation and the environ-mental characteristics of the territory surrounding any spe-cific site undoubtedly had a great deal to do with thetypes of foods which were available. In addition, thestrategies which people used to hunt, fish, gather wildplants. consume and possibly store various food items. aswell as the site specific conditions of organic preserva-tion will affect the types of food remains which mighthave survived at Early woodland residential sites. Muchremains to be learned.

Ideology and Mortuary Practices

The Early Woodland period in Vermont is best knownfrom mortuary sites, most of which were accidentallyencountered in the nineteenth or early twentieth centurywhile quarrying for sand and gravel or by constructionrelated activities. Early Woodland burials may consist ofindividual graves, but burials are recognized more com-monly when larger Middlesex cemeteries are encoun-tered, due to the presence of red ochre, copper beads andother burial items. There are four mortuary centers orcemeteries in the Champlain lowland attributable to theMiddlesex mortuary complex: Boucher (VT-FR-26),Swanton, also known as the Frink Farm or the Hemp Yard(VT -FR -1), East Creek (VT -AD- 26) and Bennett (VT -AD-298). Based on the reported presence of one or moreitems found almost exclusively in burial contexts, partic-ularly blocked-end tubular pipes and birdstones, at leasttwo more sites at the lower end of the Missisquoi River(VT-FR-16 and VT-FR-48), one site in the lowerWinooski watershed in Burlington, and a site inWoodstock in the Connecticut watershed in easternVermont may have also contained individual burials orsmall Middlesex cemeteries (Crisman 1981: 10, 81;Gifford 1948; Haviland and Power 1981; Loring 1985;Perkins 1873; Perry 1868).

Of the four better recorded cemeteries, the Boucher site isthe most completely documented (Heckenberger,Petersen and Basa 1990). Discovered by accident in 1973,the Boucher site contained 84 pit features which wereclearly burials (based on the presence of human boneand/or artifacts) and 18 additional pits which are pre-sumed to have contained human remains which havetotally disintegrated. Graves varied from basin-shapedpits less than one meter in depth to conical-shaped pitsover two meters deep. In all, the cemetery is believed tohave contained over 100 individuals. Based on a seriesof 24 radiocarbon dates ranging between 2835 and 2050B.P., it is clear that the Boucher cemetery was usedthroughout the Early Woodland period. The Swanton,East Creek and Bennett sites are undated.

Burials included both burned and unburned skeletalremains of all ages and both sexes. Most graves containedsingle individuals, but in xorne ca<:es, several peoplewere buried together. Some individuals were buried short-ly after death in flexed positions on their sides. In otherinstances, the flesh was removed from the bones, eitherintentionally or through natural decay, and the disartic-ulated bones were buried in bundles or bags, in bark con-

72

tainers or laid on bark mats. Some cremations were alsowrapped in bundles, while others were not. Over half of theskeletons were covered with red ochre, while yellow ochrewas used occasionally. This variability in burial practicesis typical of other cemeteries throughout the Northeastattributable to the Middlesex complex.

It is not known why such variability exists, but it has beensuggested that individuals who died locally during thewarmer seasons of the year were buried in the flesh, whilebundle burials and cremations may be the result of havingto transport the deceased to the cemetery from a consider-able distance or of having to wait until the ground thawedin the spring before burial (Heckenberger, Petersen andBasa 1990:211-212). The presence of a crematory at theactual site of the cemetery on East Creek suggests thatother considerations may have also been involved (Loring1985:100).

Due to the presence of copper salts in many burials, humanskeletal remains are sometimes well preserved. As withthe perishable hide, textile and fiber artifacts in thesegraves, human remains can provide a rare look at theconditions of human life which cannot be derived in anyother way. For example, limited analyses of the humanremains from the Boucher cemetery suggest that infantmortality was as high as 25% and that it was unusual forindividuals to live much beyond the age of 50. Whilethere was a low incidence of cavities, the crowns of peo-ples' teeth were often ground flat by middle age andbone surrounding some teeth became abscessed due toinfection. Broken bones were not always reset properlycausing subsequent malformation of limbs and joints.Osteoarthritis produced back problems for some mid-dle aged individuals. Possible examples of bone canceror infection and evidence of systemic osteomyelitiswhich suggests a general infection throughout the bodyconfirms that life for some individuals was not easy(Krigbaum 1990).

For all the study which has occurred, little is knownabout the specific spiritual beliefs underlying theMiddlesex burial complex. The inclusion of "medicine"bags and other unusual remains such as animal feet in sev-eral graves strongly suggests a belief in spirit helpers dur-ing one's lifetime, and perhaps during an afterlife.Artifacts such as pots and blocked-end tubes were inten-tionally broken or ritually "killed" at the time of burial,suggesting that inanimate objects contained spirits orsouls which could be released to accompany the deadindividual. It is unclear whether people in Vermont shared


a singular belief system with communities throughout theNortheast, but widespread communication among contem-poraneous Early Woodland populations seems likely. Thebroad regional exchange patterns which funneled cop-per, shell and other artifacts into Vermont would haveprovided at least one mechanism for the frequentexchange of ideas. Clearly, much remains to be learned,but any future studies must be done in close consultationwith Vermont's Abenaki community which has verystrong feelings about the disturbance of any burials, inad-vertent or otherwise.

Early Woodland Period: References

Belcher, W. R.1989 Prehistoric Fish Exploitation in East Penobscot Bay,Maine: The Knox site and Sea-Level Rise. Archaeologyof Eastern North America, Vol. 17:175-190.

Bolian, C. E. and J. B. Gengras1991 Report on Phase III Data Recovery Excavations atthe Stewartstown-Canaan Bridge Replacement Project.Prepared for the New Hampshire Department ofTransportation.

Crisman, K. J.1980 The Lower Missisquoi River Region: AnArchaeological Investigation. Honors thesis, Departmentof Anthropology, University of Vermont, Burlington.

Funk, R. E.1976 Recent Contributions to Hudson Valley Prehistory.New York State Museum and Science Service Memoir 22.

Gifford, S. M.1948 A Brief Summary of Three Years' Digging on theOrwell Site. The Bulletin of the Fort Ticonderoga Museum,Vol. 8 (1):26-28.

Haviland, W. A. and M. W. Power1981 The Original Vennonters: Native Inhabitants, Pastand Present. Hanover: University Press of New England.

Heckenberger, M. and J. B. Petersen1988 Archaeological Investigations at the SkitchewaugSite: A Multicomponent Stratified Site in Springfield.Windsor County, Vennont. Report prepared for theVermont Division for Historic Preservation,Archaeological Research Center, University of Maine atFarmington, Farmington, Maine.

73


Heckenberger, M. J., J. B. Petersen and L. A. Basa1990b Early Woodland Period Ritual Use of PersonalAdornment at the Boucher Site. Annals of the CarnegieMuseum, Vol. 59(3): 173-217.

Krigbaum, J.1989 Subsistence and Health in an Early WoodlandSkeletal Population from Vermont. Society for AmericanArchaeology Abstracts, 54th Annual Meeting, April 4-9.

1990 The Skeletal Remains from the Boucher Site.Northeast Anthropological Association Abstracts, AnnualMeeting, March 28-31, Burlington, Vermont.

Loring, Stephen1985 Boundary Maintenance, Mortuary Ceremonialismand Resource Control in the Early Woodland: ThreeCemetery Sites in Vermont. Archaeology of EasternNorth America, Vol. 13:93-127.

Perkins, G. H.1873 On An Ancient Burial Ground in Swanton,Vermont. Proceedings of the American Association for theAdvancement of Science, Vol. 22:76-100.

Perry, J. B.1868 On the Swanton Site. Proceedings of the BostonSociety of Natural History, Vol. 23 (3):247-254.

Petersen, J. B. and N. Hamilton1984 Early Woodland Ceramic and Perishable FiberIndustries from the Northeast: A Summary andInterpretation. Annals of the Carnegie Museum, Vol. 53(14):413-445.

Petersen, J. B., M. J. Heckenberger and P. A. Thomas1988 Archaeological Investigations at the Pearl StreetPark Sites in Essex, Chittenden County, County, Vermont.Department of Anthropology, Report 95. University ofVermont, Burlington.

Ritchie, W. A1969 The Archaeology of New York State [revised edition].Garden City: Natural History Press.

1970 Archeological Site Survey ill Western Vermont in1970. Report to Wenner-Gren Foundation forAnthropological Research on Grant No. 2562. Ms onfile, Vermont Division for Historic Preservation.

Smith, B. D.

1989 Origins of Agriculture in Eastern NorthAmerica. Science, Vol. 246:1566-1571.

Thomas, P. A and W. A Bayreuther1979 Chase Mill Hydroelectric Project: Phase I andII Archaeological Assessment. Department ofAnthropology, Report 13. University of Vermont,Burlington.

Thomas, P. A, G. Campoli, P. Doherty and J. B. Petersen1981 Phase I Archaeological Assessment for theMissisquoi Hydro-electric Project: Sheldon Springs,North Sheldon, Enosburg Falls North Troy and Jay BranchElements. Department of Anthropology, Report 34.University of Vermont, Burlington.

Thomas, P. A and R. S. Dillon1985 The Highgate Converter Station Site [VT-FR-161]. Department of Anthropology, Report 55.University of Vermont, Burlington.

Thomas, P. A., P. Doherty and R. S. Dillon1983 A Cultural Resource Reconnaissance for theConstruction and Operation of the Missisquoi RiverFlood Control (and Hydro-electric) Project in Richford,Vermont [VT-FR-156-157}. Department of Anthropology,Report 43. University of Vermont, Burlington.

MIDDLE WOODLAND PERIOD(2050-900 B.P.)

Regional Context

In Vermont, Middle Woodland period sites are fairly com-mon and radiocarbon dates help to anchor various com-ponents in time. Due to excavations within several largesites in the Champlain Lowland, some aspects of theMiddle Woodland cultural system are better documentedthan they are for all other periods of prehistory.

The Winooski site, located on the east bank of theWinooski River about a half mile downstream from thetirst falls and rapids, has become the type site for under-standing the Middle Woodland sequence and artifactassemblages in western Vermont (Petersen and Power1983; Figure 5). Occupation components dating after thefirst century of the Christian era were encountered in aseries of stratified alluvial deposits, allowing discreteassemblages of artifacts from different periods to beidentified and analyzed. Based on detailed studies, partic-ularly of the technological and stylistic changes in

74


Figure 5. Middle Woodland period sites cited in the text.

1. VF-FR-72. Mudgett Island3. Bessette 14. VF-CH-2015. McNeil5. Winooski6. McNeil Borrow7. VF-eH-1278. Palmer9. Ewing10. VF-WA-3511. Rivers12. Leicester Flats13. Sutherland Falls14. Skitchewaug

0 10 20 :10 r eot--=i I I IMILES

0 10 0 :10 40 0

KILOMETERS

75


ceramic vessels, Petersen defined three distinct MiddleWoodland phases between approximately 2050 and 900B.P.: the Winooski phase, the Intervale phase, and theColchester phase. A stratigraphic gap existed between theWinooski and Intervale phase components at theWinooski site.

Close analogs to all these phases occur at sites to the northin the upper St. Lawrence drainage of southern Quebec, tothe west in the Lake Ontario drainage of New York and theProvince of Ontario, and to the south in the Hudsondrainage of eastern New York (Clermont and Chapdelaine1978; Ritchie 1969; Ritchie and Funk 1973). Because ofthe broad regional cultural continuity reflected in thearchaeological record during the entire Middle Woodlandperiod, the concept of the Point Peninsula tradition hasbeen adopted to reflect this commonality. The PointPeninsula tradition has been subdivided into phases toaccount for changes in technology and other aspects of thecultural systems during this 1,000 year period. In NewYork, these include in chronological order the CanoePoint phase; the Fox Creek, Kipp Island and Burnt Hillphases; and the Hunter's Home phase (Petersen and Power1983:407-438).

Technology

Winooski Phase

The Winooski phase is defined at the Winooski site byceramics recovered from Cultural Level III, with associat-ed dates of 1790 ± 130 and 1780 ± 155 B.P. Its equivalentto the north and west is the Canoe Point phase, which,based on a number of radiocarbon samples, is estimated todate between roughly 2050 and 1650 B.P. This period ispoorly known in the Hudson valley and farther south. TheWinooski phase assemblage of flaked stone artifacts is fair-ly limited at the Winooski site. A single projectile pointdescribed elsewhere as Oxbow Type A is perhaps diagnos-tic of this period. These points are typically made fromflakes which have been bifacially reduced to form anelongated blade with weak shoulders and a contractingstem. Other tools consist of retouched flakes, end scrap-ers and a wedge. Ground stone artifacts include an abrad-ing stone, a celt, and an unfinished celt-shaped implementof greenstone. Two small, rolled copper beads are appar-ently associated with this assemblage (Petersen andPower 1983:376-377, 380).

Predominantly non-local materials were used for the man-ufacture of flaked stone artifacts. In the Winooski site

assemblage, 75% of the tools are "exotic grey chert, withlesser amounts of local black chert, exotic grey mottledchert and exotic green chert." Of the flakes, 61% are"exotic grey chert, with lesser amounts of [local] greyquartzite, red jasper [probably Colchester jasper] andwhite quartz" (Petersen and Power 1983). These highfrequencies of non-local materials, in addition to the pres-ence of copper beads, provide strong evidence for the exis-tence of long distance exchange networks extending far tothe west and south. Ceramic vessels were made with a claycontaining grit temper using a coiling technique.Vessels are cylindrical, surfaces are generally smoothed,and the firing technique is generally of high quality andis superior to that of later phases of the Middle Woodlandperiod. Body decoration was produced with a numberof stamps and possibly incising tools which were pressedinto the clay at different angles. There was a clear pref-erence for all-over decoration extending from the lip tothe base. Pseudo scallop shell decoration is present on allvessels.

Fox Creek Phase

At the Winooski site, there was no evidence of an occupa-tion dating between ca. 1650 and 1450 B.P., although thisperiod may be represented at VT-CH-20l located a fewmiles downstream on a small tributary of the WinooskiRiver. In eastern New York, this period would general-ly be equivalent to the Fox Creek phase of the PointPeninsula tradition.

At VT-CH-201 the artifact assemblage is similar to but notas extensive as that recovered at Fox Creek-related sites inthe Hudson drainage. Flaked tools consist of a FoxCreek Stemmed projectile point, a chert preform, frag-ments of two chert drills and two utilized chert flakes.All ceramic vessels are grit-tempered and exhibit a fairlywide range of decorative motifs and applications. Dentatedecoration is most common and was applied with simple,drag and rocker stamping techniques. Decoration pro-duced with a pseudo-scallop shell tool is also well repre-sented (Dillon, Thomas and Doherty 1985:38-53).

Intervale Phase

The Intervale phase was identified at the Winooski site byceramics recovered from Cultural Level II, with associat-ed dates of 1295 ±150, 1225 ±135 and 1200 ±130 B.P.Based on similarities in ceramic assemblages from othersites, it appears to be contemporaneous with theFourmile phase of the central Hudson drainage and the

76

Burnt Hill phase in the northern Hudson River and LakeGeorge area. The Intervale phase has clear analogs withthe Kipp Island phase in the upper St. Lawrence, LakeOntario and upper Susquehanna drainages. From radio-carbon dates obtained from a few sites, the temporal rangeis estimated to be roughly 1450-1150 B.P.

The Intervale phase is characterized at the Winooski siteby a relatively large and diverse flaked stone assemblage.Jack's Reef corner-notched projectile points and pointfragments were the most common (83%), although lance-olate Jack's Reef pentagonal, triangular Levanna, andside-notched points categorized as Long Bay were alsorecovered. Flake tools are a dominant artifact categorywhich consists predominantly of flakes with preparedcutting edges, end scrapers, bifacial scrapers, wedges, andan expanding base drill. Non-chipped artifacts include apendant fragment and possibly a barbed bone point. Atthe nearby McNeil site, a contemporaneous assemblageincludes Jack's Reef Corner-Notched points, a corner-notched point reworked into a drill, a small chert endscraper, a large utilized chert flake, and two fragments of aclay pipe bowl.

At the Winooski site, raw materials are varied and derivefrom both non-local and local sources. A similar pattern isevident at the McNeil site across the Winooski Riverwhere the lithic debitage associated with an Intervalephase assemblage consists almost exclusively of over1,900 exotic grey chert and 79 brown jasper flakes(Thomas and Bourassa 1978). The long-distanceexchange of raw materials brought in via networks extend-ing to the south and west and to the east into northern NewHampshire and northeastern Maine may have facilitatedcultural interaction among peoples living throughout awide geographical area.

Ceramic vessels were probably cylindrical. Most vesselsexhibit a very poor quality of firing and have disintegratedbadly at most sites. Body decoration was produced withwavy line, dentate, punctate, a variety of perishable fiberstamps, and trailing/incising tools. The presence of dis-tinctive wavy line decoration may be an especially diag-nostic marker for sites of the Intervale phase.

Colchester Phase

The Colchester phase was identified at the Winooski siteby ceramics recovered from Cultural Level I, dated to1100 ±135, 1050 ±125, and 885 ±130 B.P. Based on com-parisons with ceramic and other assemblages sites in both

The Journal ofYerrnont Archaeology

the Hudson, as well as the upper St. Lawrence and LakeOntario drainages, this phase has clear analogs with theHunter's Home phase in this region. Radiocarbon datesrelated to Hunter's Home phase components in theNortheast generally range between 1100 and 950 B.P.

The Colchester phase assemblage of flaked stone toolsat the Winooski site contains a fairly narrow range of arti-facts. The sole projectile point form is represented by 38triangular Levanna points. Other flaked tools include 70retouched flakes or prepared end scrapers, one pentagonalbiface, one drill and four wedges. A sandstone abradingstone was the only recovered tool related to a groundstone industry. A more widely varied tool assemblageat the multicomponent Rivers site at the mouth of DeadCreek in the Otter Creek valley includes bone or antlerfishhooks, gorges, awls and needles (Bailey 1939; Basa1975).

Unlike earlier phases, lithic raw materials used forflaked tools were predominantly local in origin. At theWinooski site, local quartzite and black chert are repre-sented by approximately 85% of both the tools and wasteflakes. Across the Winooski River at both the McNeil siteand the McNeil Borrow site, the latter dated to 1010 ±11OB.P., a similar pattern is evident. Here, the Colchesterphase assemblages reflect the exclusive use of localquartzite and Champlain valley chert (Thomas andBourassa 1978; Thomas and Robinson 1981). This shiftfrom non-local to local raw materials appears to reflecta significant breakdown in long-distance exchange net-works, and possibly the establishment of a culturalboundary between local populations and those of theHudson drainage and farther west (Petersen and Power1983:453).

Ceramic vessels were cylindrical and bulbous with slight-ly pointed to rounded bases. The principal form of dec-orative application was simple vertical stamping whichwas restricted to the upper neck, rim, lip and upper inte-rior portions of each vessel. The vast majority of theassemblage is decorated with oblique or horizontal cord-wrapped stick or complex cordage impressions. Circularpunctations are also a common decorative technique. Asmall number of vessels are entirely undecorated.

In sum, the general characteristics of flaked stone andceramic assemblages dating to the Middle Woodlandperiod are reasonably well known, with the exception ofthe early Middle Woodland period, which is representedat only a few sites. Intervale and/or Colchester phase

77


assemblages at other sites on the Winooski intervale, onthe Missisquoi delta and upstream along the MissisquoiRiver in Highgate, on Shelburne Pond and at the mouth ofDead Creek all tend to mirror similar patterns withrespect to the types of projectile points encountered,changes in the use of various types of raw materials,and the techniques used to decorate ceramic vessels (Bailey1939; Petersen et al. 1983; Thomas 1989; Thomas andDillon 1983; Thomas and Robinson 1979). Roughlycomparable Middle Woodland period assemblages arealso likely to be present in the Connecticut River water-shed of eastern Vermont (Heckenberger and Petersen1988:90-113).


After a thousand years of deteriorating conditions, the cli-mate seems to have stabilized somewhat about 1800 B.P.Subsequently, a slight warming trend occurred. It issuggested that this stabilization led to a chain of eventswhich had direct implications for the nature of human set-tlement and subsistence patterns adopted during the Middleand early Late Woodland periods. A much higher inci-dence of Middle Woodland period sites relative to theEarly Woodland period may also signal a significantpopulation growth throughout the region during this thou-sand year period.

Beginning about 1650 B.P., the Champlain Lowlandapparently experienced a gradual population increase.Based on current information, sites of this period are locat-ed exclusively in the Champlain valley and along theConnecticut River. By 1200 B.P., extensive settlementscould be found in the lower reaches of all major rivers(Power, Cowan and Petersen 1980; Thomas and Doherty1981). It is suggested that such an apparent populationgrowth would have been sufficient to have overtaxed thenatural resources had these groups remained exclusivelyin the lowlands. A new diversified settlement patterngradually evolved. During some seasons, groups may havecongregated in lowland areas where fish, plant and gamespecies were dense and varied. At other seasons, fairlylarge communities may have broken up into small bandsor employed specialized task groups to better exploitdispersed food resources, including those in the uplands.By roughly 1000 B.P., groups are likely to have consis-tently utilized portions of all of Vermont's watersheds aslowland populations once again intensively used theuplands of Vermont, at least on a seasonal basis. Withsome variation, including the introduction of agricultureduring the Late Woodland period, this expanded pattern

probably continued until the arrival of Europeancolonists in the seventeenth century.

Winooski Phase

The settlement pattern of the Winooski phase can beinferred only in part from a small but varied sample of sitelocations. A variety of riverine and lacustrine associationsoccurs, in many cases in close proximity to extensivewetlands. Precise settlement data, including the spatialextent of occupation, are very limited. The configurationof house structures is unknown. Initial estimates of set-tlement size, which are merely probable maximumextents of site utilization, range from a maximum sizeof 405 square meters for the Weinman site on Lake Georgeand 400 square meters for the Ewing site on ShelburnePond to 135 square meters for the Otter Creek No. 2 siteon Otter Creek. Along with the data from the Winooskisite, these estimates help to confirm the small size of res-idential sites of the Winooski phase. Presumably, sitesof this phase reflect the presence of a restricted wan-dering community pattern in the Lake Champlain drainageduring the period (Petersen and Power 1983:396).

At the Winooski site, food remains were recovered from13 hearths associated with probable late summer to falloccupations during the early Winooski phase. Animal andplant remains include deer, domestic dog, numerousfragments of unidentified mammal bone, a fragment ofbird bone, unidentifiable fish bone, as well as butternut,pigweed, fire cherry and bedstraw. Faunal and floralremains recovered at other sites cannot be reliably associ-ated with the Winooski phase. Similarly, no estimates ofseasonality are available (Petersen and Power 1983).

Fox Creek Phase

Settlement data for the tentative Fox Creek phase aresparse, as only a few sites are known. Residential sitelocations include VT -FR-7 on the lower Missisquoi Riverand VT-CH-201 situated on a delta terrace above theWinooski River several miles up stream from LakeChamplain, while a small activity site may be representedby VT-CH-127 located on a narrow terrace below theWinooski Falls. A radiocarbon date of 1570 +/-70 B.P.suggests a roughly contemporaneous residential occupa-tion at the Skitchewaug site on the Connecticut River inSpringfield (Crisman 1980:23-34; Dillon, Thomas andDoherty 1985:38-53; Heckenberger and Petersen 1988;Thomas and Bayreuther 1979).

78

Site size is unknown' and can only be approximated atVT-CH-201. Here, a Fox Creek point and associatedceramics were recovered from a focal activity area esti-mated to be approximately 400 m2. House types areunknown. It seems reasonable to assume that the settle-ment pattern characteristic of the Fox Creek phase was notsubstantially different from that of the earlier Winooskiphase.

Subsistence data related to the tentatively assigned FoxCreek phase are uncommon from excavated sites.Suspected site locations, such as the alluvial terracesabove the Salmon Hole below Winooski Falls, LeicesterFlats, Mudgett Island and at Sutherland Falls, would seemto indicated a focus on riverine and wetland resourcesduring at least part of the year. Bones recovered inassociation with several hearths at VT-CH-201 locatedabove the floodplain in the lower Winooski watershedare cliderived from both the lower legs and skull of one ormore deer, while a tooth fragment may derive fromblack bear. Deer antler was also identified (Dillon,Thomas and Doherty 1985:49-50). The season of occupa-tion is unknown.

Intervale Phase

The settlement pattern of the Intervale phase can be par-tially reconstructed, based on the association of overtwenty sites with a variety of environmental settings. Incommon with the earlier Winooski and Fox Creek phas-es, small residential camps or base camps are locatedin a number of riverine and lacustrine microenvironments,often in close proximity to a wetland. Most known sitesare located on floodplains along the lower 25 miles of theMissisquoi, Lamoille and Winooski Rivers and OtterCreek, but other sites are situated on the shores of LakeChamplain and Lake George, on Bristol Pond andShelburne Pond, and on smaller rivers such as theLeicester River. What is probably a small stop-over site,'VT-WA-35, is situated on a narrow terrace along theWinooski River in Moretown. A single hearth contain-ing a few fragments of burned bone dates to 1450 +/-90B.P. This sample of sites undoubtedly reflects a variety ofseasonally occupied stations within an annual settlementsystem which might be characterized as a restricted wan-dering or central based wandering community pattern(Petersen and Power 1983: 402-403; Thomas 1989).

Estimates of settlement size can be made from a limitedsample of Intervale phase sites. Estimates for very smallto small residential camps range between approximately 56

<IIIEIII•..~U<II

1'1III

'"Z0 "um 30zC(

<IIIIIIE::::l!;(III.....J..a::::>•....J:::>o

H

o

".Ia:u,I•..>

79


!

•

.'-0

" -:• I

~

H


and 500 square meters. Several of these camps presum-ably represent small extractive or processing stations with-in the settlement system. Estimates of very large residen-tial areas which probably represent recurrently occupied,summer-fall base camps range between roughly 5,400square meters for the Reynolds site on the MissisquoiRiver up to 9000 square meters at sites located in theMissisquoi delta. These estimates, along with those forLocus 1 and possibly Locus 5 at the Winooski site, sub-stantiate the existence of variably sized settlements duringthe Intervale phase. Such variation presumablyreflects the seasonal aggregations and dispersal of fami-lies in a central based wandering community pattern(Petersen and Power 1983:403).

Food remains have been recovered at a number of siteswhich contain components related to the Intervale phase.At the Winooski site, a wide spectrum of plants and ani-mals were utilized during the summer and into the fall.Animal remains include burned bone derived from 33hearths, including deer, beaver, muskrat, woodchuck,chipmunk, domestic dog, lake sturgeon, bullhead, as wellas many fragments of unidentifiable mammal, fish andbird bone. Plant remains include butternut, black walnut,hickory nut, hazelnut, pigweed, blackberry, staghornsumac, mustard, hog peanut, chairmaker's rush, snowber-ry, elderberry, sassafras, buckwheat, and bedstraw. At theMcNeil site across the river, nearly 7,000 fragments ofburned bone, mostly shattered long bone, were recoveredfrom a large hearth. Identifiable species include wood-chuck, squirrel and deer (cranial and leg bones). Bothsites appear to represent warm season occupations.Upstream, the presence of butternut at VT-CH-127, locat-ed near the Winooski Falls, provides evidence of anautumn occupation (Petersen and Power 1983; Thomas andBayreuther 1979; Thomas and Bourassa 1979).

A diversified food base is also suggested by data collectedfrom sites along the Missisquoi River to the north.Bullhead was identified in a hearth at what is probably alarge base camp at the lower end of the river. Fartherupstream in Highgate at the Bessette 1 site, a large hearth,associated with both Jack's Reef Pentagonal andLevanna projectile points and dated to 1180 +/-90 B.P.,contained almost 2,000 fragments of burned bone (Figure6). Deer is the most common animal represented.Bones are predominantly from the lower legs, althoughfragments of upper leg, vertebrae, scapulas and teeth arealso present. Other species include squirrel, bear, beaver,and possibly fox or raccoon. Nut fragments were alsorecovered (Thomas and Dillon 1983; Thomas and

Robinson 1979).

Data on seasonality and subsistence systems seem heav-ily weighted towards summer-fall periods of occupationat Intervale phase sites which produced subsistenceremains, although an extended spring-late fall occupa-tion may be represented at the Bessette 1 site on theMissisquoi River. Sample size bias and the lack of coldseason indicators may well account for this impression.From what is currently known, a relatively diffuse subsis-tence economy based on hunting, fishing and gatheringseems apparent for the Intervale phase.

Colchester Phase

The settlement pattern of the Colchester phase is hypoth-esized on the basis of the environmental settings of numer-ous but mostly undated sites assigned to this period. Sitesare present in a wide variety of settings. Large sites aregenerally situated on the floodplains along the lowerWinooski, Lamoille and Missisquoi Rivers and along thelower stretches of Otter Creek. Other sites occur alongthe middle and upper reaches of these rivers, alongnumerous interior streams, such as the Leicester, Hoosicand Walloomsac Rivers, on even smaller streams, such asMuddy, Potash and Sunderland Brooks, on bluffs over-looking the valley bottom, and around the shores ofShelburne Pond, Lake St. Catherine and Lake Bomoseen.This sample of sites represents some portion of one ormore annual settlement systems, with seasonal occupa-tion clearly indicated in some instances (Petersen,Heckenberger and Thomas 1988:129-134; Petersen andPower 1983:405-406; Thomas and Bayreuther 1980;Thomas, Campoli and Doherty 1979; Thomas and Dillon1985; Thomas and Kochan 1987; Thomas andMatthews 1986).

Settlement size can be approximated for sites of theColchester phase in some cases. Estimates range from amaximum area of 400 m2 at the Ewing and Palmer sites onShelburne Pond and Potash Brook to at least 6,500 m? atthe McNeil site on the Winooski Intervale. These esti-mates, used in conjunction with data from the Winooskisite, substantiate the presence of three kinds of sites:large aggregated settlements, undoubtedly occupied duringthe warmer seasons; smaller, possibly cold weather resi-dential camps; and what appear to be primarily extrac-tive (hunting, fishing or gathering) camps located aroundinland ponds and small streams. These data indicate thecontinued existence of a central based wandering commu-nity pattern, but with the use of a greatly expanded territo-

80

1-ry as exploitation of the uplands became increasinglycommon. By the end of the Middle Woodland period inthe Champlain drainage, the settlement pattern may havebegun a shift to a semipermanent sedentary communitypattern. Such a trend remains to be demonstrated (Petersenand Power 1983:406).

Faunal and floral samples recovered from 13 Colchesterphase hearths at the Winooski site indicate a fairly widesummer to fall subsistence base, including deer, fox, lakesturgeon, bullhead, unidentified bird and turtle, butternut,acorn, pigweed, rose, mustard, buckwheat and bedstraw.At the nearby McNeil site, enamel from a beaver incisor, afragmentary bird bone and unidentifiable burned mam-mal bone were recovered from a feature, with possiblewarm season implications. Although of more equivocalassociation, a wide variety of subsistence remains havebeen recovered from the Rivers site at the junction of Deadnd Otter Creeks. Represented species include deer, bear,moose, elk, beaver, muskrat, porcupine, dog, fisher,perch, bullhead, and unspecified mammal, bird, fish andturtle. These remains seemingly indicate varied seasons ofoccupation, presumably centered on the warm sum-mer/autumn months. A comparable admixture of subsis-tence remains was obtained from the Ewing site, includ-ing deer, bear, moose, beaver, muskrat, fisher, snappingturtle, bullhead, and unspecified mammal, bird, fish andturtle, as well as butternuts and acorns. Again, these sug-gest several seasons of occupation in the warm months ofthe year.

Subsistence and seasonality data suggest a series of warmseason occupations for components of the Colchesterphase. However, the wide range of subsistence remainsrecovered and the likely bias produced by a lack of coldseason indicators prevents assignment of these sites to sole-ly warm season occupations. A diffuse pattern of hunt-ing and gathering, on the other hand, is well docu-mented for the Colchester phase.

A great deal of variability may actually exist in the typesof food remains which are likely to be encountered atMiddle Woodland period sites. The presence of broad ornarrow ranges of plant and animal remains may resultfrom such factors as seasonality of site occupation,the environmental characteristics of the territory surround-ing any specific site, and the specific types of techniqueswhich people used to hunt, fish, gather, consume andpossibly store various food items. There is still much to belearned.


Ideology

Little evidence related to the spiritual beliefs of peopleduring the Middle Woodland period has been discoveredat sites in Vermont. Although evidence of elaborate bur-ial ceremonialism is evident to the west in central and west-ern New York, particularly during the Squawkie Hill andKipp Island phases, there is no evidence to suggest suchpractices here.

Middle Woodland Period: References

Bailey, J. H.1939 Archaeology in Vermont 1938. In Archaeologyin Vermont [revised edition], edited by 1. C. Huden, pp.3-6. Rutland, VT: Charles Tuttle, Co

Clermont, N. and C. Chapdelaine1978 Une Station Cosmopoloite du Sylvicole Moyen:Pointe-du-Buisson No.3. Canadian Journal ofArchaeology, Vol. 2:79-100.

Crisman, K. J.1981 The Lower Missisquoi River Region: AnArchaeological Investigation. Honors thesis, Departmentof Anthropology, University of Vermont, Burlington.

Dillon, R. S., P. A. Thomas and P. Doherty1985 Archaeological Reconnaissance Survey and SiteEvaluation Study for the Yellow-Blue-Yellow Alignment,Chittenden County Circumferential Highway, 1984-1985.Volume V: Technical Data for Sampling Areas andArchaeological Sites. Department of Anthropology,Report 65. University of Vermont, Burlington.

Heckenberger, M. and J. B. Petersen1988 ArchaeologicalInvestigations at the SkitchewaugSite: A Multicomponent Stratified Site in Springfield,Windsor County, Vermont. Report prepared for theVermont Division for Historic Preservation,Archaeological Research Center, University of Maine atFarmington, Farmington, Maine.

Petersen, J. B., M. J. Heckenberger and P. A. Thomas1988 Archaeological Investigations at the Pearl StreetPark Sites in Essex. Chittenden County. County. Vermont.Department of Anthropology, Report 95. University ofVermont, Burlington.

Petersen, J. B. and M. W. Power1983 The Winooski Site and the Middle Woodland Period

81


in the Northeast. Report submitted to InteragencyArchaeological Services, Mid-Atlantic Region, NationalPark Service.

Petersen, 1. B., J. A. Wolford, N. D. Hamilton, L. LaBarand M. Heckenberger1985 Archaeological Investigations in the ShelburnePond Locality, Chittenden County, Vermont. Annals ofCarnegie Museum, Vol. 54 (3):23-75.

Power, M. W, F. L. Cowan and J. B. Petersen1980 Artifact Variability at the Multi-componentWinooski Site. Man in the Northeast, Vol. 19:43-55.

Ritchie, W. A.1969 The Archaeology of New York State [revised edia-tion]. Garden City: Natural History Press.

Ritchie, W. A. and R. E. Funk1973 Aboriginal Settlement Patterns in the Northeast.New York State Museum and Science Service, Memoir 20.

Thomas, P. A.1980 The McNeil Generating Plant Site (VT-CH-93),Burlington, Vermont. Man in the Northeast, Vol. 19:57-71.

1989 Phase 1A Archaeological Reconnaissance Survey,Waterbury-Duxbury-Moretown BRF 013-4(10),Washington County, Vermont [VT-WA-35). ConsultingArchaeology Program, Department of Anthropology,Report 101. University of Vermont, Burlington.

Thomas, P. A. and W. M. Bayreuther1979 Chase Mill Hydroelectric Project: Phase I and IIArchaeological Assessment [VT-CH-127 and VT-CH-128}. Department of Anthropology, Report 13.University of Vermont, Burlington.

1980 Wells RS 0145(7): Phase I ArchaeologicalAssessment. Department of Anthropology, Report 25.University of Vermont, Burlington.

Thomas, P. A. and M. L. Bourassa1978 McNeil Generating Plant: Phase II IntensiveArchaeological Survey [VT-CH-93]. Department ofAnthropology, Report 2. University of Vermont,Burlington.

Thomas, P. A., G. Campoli and P. Doherty1979 An Initial Study of the Prehistory and Historic

Archaeological Sensitivity ofDepartment of Anthropology,Vermont, Burlington.

the Bennington Area.Report 22. University of

1983 Intensive Archaeological Evaluation for SelectedPrehistoric Sites within the Highgate FallsHydroelectric Project Area and PrehistoricArchaeological District [VT-FR-103, 104, 134, 139, 140).Department of Anthropology, Report 44. University ofVermont, Burlington.

Thomas, P. A. and R. S. Dillon1983 Intensive Archaeological Evaluation for SelectedPrehistoric Sites within the Highgate FallsHydroelectric Project Area and PrehistoricArchaeological District. Department of Anthropology,Report 44. University of Vermont, Burlington.

1985 The Highgate Converter Station Site [VT-FR-161]. Department of Anthropology, Anthropology,Report 55. University of Vermont, Burlington.

Thomas, P. A. and P. Doherty1981 Archaeological Resource Management Study,Lower Winooski River Watershed. Department ofAnthropology, Report 37. University of Vermont,Burlington.

Thomas, P. A. and G. P. Kochan1987 Archaeological Reconnaissance Survey forWallingford RS 0137(5), Wallingford, Vermont [VT-RU-158). Department of Anthropology, Report 78. Universityof Vermont, Burlington.

Thomas, P. A. and W. Matthews1986 Archaeological Reconnaissance Survey, PalmerDevelopment Parcel, Kimball A venue, South Burlington[VT-CH-255]. Department of Anthropology, Report 64.University of Vermont, Burlington.

Thomas, P. A. and B. S. Robinson1979 Missisquoi National Wildlife Refuge: A CulturalResource Survey. Department of Anthropology, Report10. University of Vermont, Burlington.

1981 JosephC. McNeil Generating Station: BorrowArea. Archaeological Evaluation [VT-CH-146).Department of Anthropology, Report 31. Universityof Vermont, Burlington.

82

LATE WOODLAND PERIOD(900-350 B.P.)

Regional Context

The Late Woodland period was a time of significantchange in cultural systems throughout New England. Thiswas both a period of population growth and territorialexpansion in the entire north Atlantic region. Norseadventurers sailed westward from the Old World toIceland, Greenland and the coast of North America.Closer to Vermont, Iroquoian-speaking communitiesfrom the Lake Ontario region apparently moved north-eastward into the lower St. Lawrence drainage. Thesecommunities were the precursors of the St. LawrenceIroquois encountered by Jacques Cartier around Montrealin 1534. Although people who spoke one or moreWestern Abenaki dialects resided along the eastern shoreof Lake Champlain and throughout Vermont, outlyingSt. Lawrence Iroquoian communities may have taken upresidence in areas along the western shore of LakeChamplain as well during at least part of the LateWoodland period. By 1607, however, all St. LawrenceIroquoian villages had mysteriously disappeared from theentire St. Lawrence watershed (Chapdelaine 1989; Martijn1990; Pendergast 1990; Petersen 1990; Snow 1980;Wright 1979).

In central and western New York, Ritchie (1969) andMacNeish (1952) have documented two, well-developed,sequential traditions defined primarily on the basis ofstylistic changes in ceramic vessels. The Owasco tradi-tion, ca. 950-650 B.P., has been subdivided into theCarpenter Brook, Canandaigua and Castle Creek phases.The Iroquois tradition, ca. 650-350 B.P., has been sub-divided into the Oak Hill, Chance and Garoga phases.Each phase lasted for approximately 100 years.

The Owasco-Iroquois sequence of central and westernNew York apparently has its analogs in the HudsonValley of eastern New York, although major Owascocomponents have not been identified here. Owasco-likeceramics in the middle and upper Hudson Valley are onoccasion accompanied by other artifacts found at Owascosites to the west; e.g., Levanna projectile points, strike-a-lights, sinewstones, scrapers, incised or punctate claypipes, large storage pits, flexed burials without gravegoods, and maize cultivation (Funk 1976:300). Artifactassemblages at sites dating after 650 B.P. also resemblethose at Iroquois sites to the west, although differencesin various attributes are evident in some of the pottery.


Despite many similarities in the material culture betweensites in the Hudson valley and sites in the Mohawk valleyand farther west, there is good reason for not using"Iroquois" as a descriptive term for sites in the HudsonValley which are less than 650 years old, because cultur-ally unrelated, Algonkian-speaking Mahicans occupiedthis valley throughout the Late Woodland period (Funk1976:302).

North of Vermont, the cultural chronology is less clear, butsubstantial strides have been made within the pastdecade in understanding the St. Lawrence Iroquois. In1979, Wright proposed, "About A.D. 1400, some of the St.Lawrence Iroquois groups probably expanded fartherdownstream to become the ancestors of the historicStadaconans [in the region around Quebec City]" (Wright1979:69). More recent work in the Province of Quebechas challenged the older idea that St. Lawrence Iroquoisdevelopment occurred exclusively west of Montreal.Chapdelaine and others have hypothesized the presence ofIroquoian groups in the region around Quebec City asearly as the Middle Woodland period and on the lower St.Lawrence at the time of European contact. The details arestill far from being worked out, however (Chapdelaine1989,1993).

Analogs of both Owasco and lroquois/St. LawrenceIroquois tradition ceramic vessels are apparently presentin the Champlain and Connecticut River drainages andelsewhere in New England. Regional diversity and ashortage of thoroughly studied and dated sites in northernNew England attributed to the Late Woodland periodmakes it difficult to relate materials from this area to spe-cific phases to the north and west. Most likely, devel-opments in the Champlain and Connecticut Riverdrainages, while related to defined phases in surroundingareas, are unique and warrant more local definition(Heckenberger and Petersen 1988).

Technology

Sites dating to the Late Woodland period occur through-out Vermont, but the actual time of their occupationhas been very difficult to determine. Radiocarbon datesexist for only six sites: Sumner's Falls (800 ± 80 B.P.),Skitchewaug (850 ± 50 to 580 ± 60 B.P.), Dewey's Mills(490 ± 120 B.P.), VT-FR-134 (510 ± 90 B.P.), VT-FR-14(}-Bessette 1 (510 ± 80B.P.) and Donohue (51O± 125and 250 ± lIS B.P.) (Figure 7). Temporally diagnosticstone tools which could be used to subdivide this period do

83


Figure 7. Late Woodland period sites cited in the text.

1. VI'-FR-1342. Bessette 13. VI'-FR-I6I4. VI'-CH-28/295. Donohue6. Shelburne Pond, Ewing7. Silver Lake8. Dewey's Mills9. Sumner's Falls10. Skitchewaug11. Bellows Falls12. Vernon Falls13. Fort Hill

0 10 20 30 40 !l0f--! Ld I II

MILES

0 10 60

KIL.OMETERS

84

Plate 7. Some artifacts from the Donohue site (VT-CH-94).a-d: Levanna points; e-h: broken bifaces; i: drill tip; j-l:end scrapers; m-n: side scrapers.

not exist, as triangular Levanna projectile points wereused from the late Middle Woodland throughout the LateWoodland period. Other stone tools show no apparentchange in form. Stylistic changes in ceramic vesselsmight provide finer chronological control, but potteryhas not been recovered at the vast majority of LateWoodland sites and a ceramic typology based on detailedattribute analyses has not been developed.

The range of lithic tool types at dated Late Woodlandsites is relatively narrow, For example, frorn a recurrently

occupied base camp (VT-FR-134) on the MissisquoiRiver, recovered lithic artifacts include Levanna projectilepoints, unifacial scrapers with steep bits, a drill, ~dchert flakes utilized as expediency tools for cutting(Thomas and Dillon 1985). At the Donohue site (VT- CH-


94), a semipermanent residential village on the lowerWinooski River floodplain, a similar range of lithic arti-facts was found, including Levanna points, a drill tip, endand side scrapers, a bifacial punch or reamer, a brokenknife, a tabular stone rod, pieces of graphite and hematite,and a large piece of crystal quartz (Thomas and Bumsted1979; Plate 7). Although not recovered from dated con-texts, additional Late Woodland artifacts include ground-stone hoes, celts, adzes, pestles, hammerstones, mulIersand grindstones.

During the first few centuries of the Late Woodlandperiod, ceramic vessels exhibit considerable stylistic conti-nuity with the preceding period. Vessels are character-istically decorated by stamping or impressed with cord-wrapped sticks. Circular punctation, incision and linearpunctation are all employed to some extent. In theConnecticut River valley, there appears to have been asignificant shift from grit to shell temper (Heckenbergerand Petersen 1988). By about 650 B.P., substantialchanges occurred in vessel form and the types of decorativetechniques employed. Globular vessels replaced the earli-er pointed- base forms. The configuration of vessel rimsbecame more elaborate. This is particularly evident onvessels similar to those of the Owasco-Iroquois tradi-tion where wide collars were invariably decorated withincised complex geometric motifs. Punctations werealso applied to collars with either fingernails or tools.Vessels are commonly thin, relatively small (ca. 1.5 gal),and very well fired (Petersen and Sanger 1990).In addition to analogous Iroquois-like vessels in theChamplain Valley, examples of distinctive Iroquois ves-sels with wide castellated rims, elaborate geometricdesigns produced by incision and dentate stamping, hol-low punctate impressions, and smoothed or checker-stamped bodies have also been found. All of these ele-ments, but particularly the dentate stamped lineardesigns and checker-stamped bodies, are particularlycharacteristics of ceramics found at St. LawrenceIroquoian sites on the lower Richelieu River which drainsLake Champlain, and at sites in the vicinity of Montreal.Thus, some researchers have proposed that a cluster of St.Lawrence Iroquois villages existed in the Champlain val-ley (Pendergast 1990). With distinct St. Lawrence Iroquoisceramic vessels present at sites as far east as westernMaine, others have suggested trade 11" an equally likelysource of origin for St. Lawrence Iroquois pottery in theChamplain valley (Petersen 1990).

Artifacts manufactured from perishable materials are rarelyfound in Vermont. A notable exception has been the

85


Plate 8. A shallow cooking hearthat the Donohue site (VT-CH-94).A carbonized corn cob was recoveredin this hearth.

recovery of at least six dugout canoes from both the low-lands and uplands. The extensive distribution and associ-ated radiocarbon dates, ranging from 510 ± 100 to 380 ± 60B.P., for dugouts in Shelburne Pond and Silver Lake indi-cate that dugouts, in addition to historically reported birchbark canoes, were frequently used in ponds and marsheswell into the Late Woodland period.

Settlement and Snbsistence

In Vermont, the first evidence of cultivation of non-indige-nous plants occurred between 850 and 500 B.P. duringthis period. Corn and bean cultivation is firmly dated toabout 510 B.P. at the Donohue site on the lower WinooskiRiver, to roughly 585 B.P. at the Hunter site, and asearly as 850 B.P. at the Skitchewaug site on theConnecticut River. The introduction of horticulture seemsto have promoted population growth and to have has-tened or at least facilitated the already established devel-op mental pattern toward larger, seasonal sedentary set-tlements.

It is perplexing that sites dating more recently than 500years ago, in contrast to those dating to the preceding threecenturies, are either uncommon or have not been general-ly recognized. For example, with the onset of the LateWoodland period, ca. 900 B.P., aboriginal utilization andoccupation of the Shelburne Pond locality increased rather

dramatically. At least 15 sites, including the Ewing site,date to this period. Little evidence of utilization of thepond after ca. 500 B.P. exists, however. Several radiocar-bon dates and a few ceramic sherds from the Ewing siteprovide the only current evidence for one possible lateroccupation (Petersen et al. 1984).

Even in the lower Missisquoi River valley where Abenakicommunities are known to have been present during theseventeenth century, an extensive review of site data ledto the identification of numerous Middle Woodland sitesalong the river, but provided little indication of occupa-tions dated to the last three centuries of the Late Woodlandperiod (Crisman 1980: 119). A similar trend is character-istic of the lower Winooski valley. Dean Snow has goneso far as to suggest that the movement of Iroquoiansinto the lower St. Lawrence valley during this period cutoff the Champlain valley inhabitants from their northernconnections, leaving the Champlain drainage a depopu-lated backwater and the Hudson-Champlain corridor abuffer zone between the St. Lawrence and Hudson Rivers(Snow 1980: 308). On the other hand, Abenaki oral tra-dition and the presence in older artifact collections ofcastellated ceramic vessels with incised and hollow-punctate designs indicate the existence of sites datingbetween 500 and 350 B.P. Research is clearly needed todetermine whether the poor visibility of such sitestoday resulted from a substantial shift in settlement focusto areas which are not commonly surveyed, from major

86

Plate 9. Cross section of a smallstorage pit at the Donohue site(VT-CH-94 ).


regional demographic changes, from site loss due to his-toric plowing and pilfering, or from other causes.

Although archaeological data related to Late Woodlandperiod sites in Vermont are very limited, a riverine-pond-lake focus apparently continued from an earlier period.Seasonal mobility was characteristic. At a general level,semipermanent settlements, base camps, small residentialcamps and extractive camps are likely site types.

By 850 B.P., it seems likely that a horticultural settlementwas active at Skitchewaug on the Connecticut River inSpringfield. What appear to be shallow pit houses withassociated storage pits were identified during salvageexcavations along slumping river banks. Carbonized cornhas been recovered from six features dated between 850and 630 B.P.; carbonized beans were retrieved from threepits dated between 850 and 760 B.P.; squash remainswere recovered from one feature dated to 830 B.P.(Heckenberger and Petersen 1988).

Archaeological remains of part of an apparent small resi-dential hamlet dating to ca. 510 B.P. were also encounteredduring excavations at the Donohue site (VT- CH-94) onthe Winooski Intervale in Burlington (Thomas andBumsted 1979:105-158). The Donohue site encompassesan area of at least 152 x 75 m (500 x 250 ft) along the west-ern edge of a relict river channel. It is likely that VT-CH-95, located along the eastern margin of the same relict

channel, is contemporaneous. If so, the larger zone ofprehistoric settlement would cover an area of roughly 110x 287 m (360 x 940 ft), or nearly eight acres.

Twenty-two cultural features and a roughly 3 x 3.5 mlithic workshop were exposed in a limited sample areawithin the Donohue site. Six hearths containing char-coal, burned bone, carbonized plant remains, some arti-facts and fire- reddened soil were identified (Plate 8).Other features included bell-shaped pits which probablyserved as storage pits (Plate 9). Possible post molds ofabout 7.5 cm in diameter were encountered in one exca-vation unit and may indicate the presence of structuralremains.

Corn in the form of both carbonized kernels and portionsof three carbonized cobs was encountered in a numberof pits, confirming the importance of domesticated cropsto the site's occupants (Plate 10). (No carbonized beanswere present as originally indicated in the report.) Inaddition to the corn, carbonized butternut shells were com-monly encountered. Other food items include acorn, grapeand unidentifiable fish. Fragments of mammal boneprobably resulted from bones being broken to extract mar-row. One unburned deer mandible may have been utilizedto scraps the kernels from corn cobs (Bumsted 1980;Thomas and Bumsted 1979:105-158). Given the pres-ence of both carbonized butternut and corn and theprobability that at least two of the storage pits were emp-

87


Plate 10. Organic food remains recovered from hearths atthe Donohue site (VT-CH-94). a: butternut shells; b: car-bonized grape pits; c-e: carbonized corn kernel; l acornshell; g: burned fish bone.

tied before they were reused for cooking, a fall-winteroccupation seems likely.

Seasonal base camps may have been located near the fallsat Vernon and Bellows Falls on the Connecticut River, atprime fishing sites along the rivers which drain intoLake Champlain, on lowland ponds and along the shoresof Lake Champlain (e.g., Bailey 1937, 1939). VT-CH-28and VT-CH-29, located close to the first falls above themouth of the Lamoille River, may represent primary fish-ing stations. Sites dating to the early Late Woodland peri-od(ca. 850-650 B.P.) around Shelburne Pond whichcontain extensive concentrations of fire-cracked rockand black midden soils and the Skitchewaug site on theConnecticut River floodplain may represent other typesof base camps (Heckenberger and Petersen 1988;Petersen et al. 1984). During the Shelburne phase (ca. 950-

650 B.P.) at the Ewing site, a wide array of animal foodswas consumed: immature and mature deer, elk or wapi-ti, bear, dog, beaver and muskrat, as well as bullhead,possibly perch, northern pike, bass, snapping turtle,unidentified birds and mollusks. Although hickory nutsand butternuts are present, no evidence of cultigens hasbeen recovered (Petersen et al. 1985). Small residentialcamps may be the most numerous type of site. They mayoccur in a wide variety of settings and may be functional-ly very different. In the Champlain lowland, early LateWoodland period sites around Shelburne Pond and nearthe headwaters of Sunderland Brook in Essex may be typ-ical of camps from which foraging activities were carriedout. Such sites contain a fairly narrow range of artifacts(Petersen et al. 1984; Petersen, Heckenberger andThomas 1988). The Chipman's Point rockshelter alongthe lakeshore may represent a travel site (Bailey 1940).Other sites in the Hoosie, Walloomsac and Jewett Brookvalleys around Bennington, at the Salmon Hole on theWest River in Jamaica in the uplands of eastern Vermont,and around such upland ponds as Lake St. Catherinemay represent small family residences in the uplands wherehunting and trapping occurred during the fall and win-ter (Thomas and Bayreuther 1980; Thomas, Campoli andDoherty 1979; Thomas, Doherty and Warren 1982:37;Thomas and Warren 1984:24-30).

Extractive camps, particularly in the lowlands, may havebeen established by specialized task groups moving outfrom villages or base camps. Low artifact densities areexpected at such sites because few people were involvedand virtually all activities were focused on the acquisi-tion and processing of a single major resource which couldbe brought back to the main settlement for consumptionand possible storage. Components within VT-FR-161 andVT-FR-140 on the Missisquoi River which contained oneor two broken Levanna points, limited evidence of toolmaintenance, and small basin-shaped hearths may repre-sent short-term extractive camps where a task groupfocused on hunting. At VT-FR-161, flotation samplesfrom two hearths produced three pieces of shagbark hick-ory nut and about 150 small fragments of unidentifiablemammal bone. Most is probably from deer bone bro-ken for marrow extraction (Thomas and Dillon 1985).

Excavations at a seventeenth-century fortified Sokoki orSquakheag village provide the only detailed artifactual,settlement and subsistence information for the end of theLate Woodland period anywhere in or near Vermont. Atthe Fort Hill site, located just across the Connecticut Riverfrom Vernon, Vermont, large quantities of animal and plant

88

__ PERIOD OF HEAVY USE

[,"';;:';·:3 PERIOD OF LIMITED EXPLOITATION

C=:J PERIOD OF LITTLE OR NO USE

Figure 8. Reconstruction of major dietary constituents inthe seventeenth-century Abenaki subsistence cycle in theConnecticut Valley.

remains were recovered from numerous refuse pits whichhad once been used for storing food. Historic documentsindicate that this site was occupied from September, 1663to perhaps the following April or May, 1664. Althoughmeat from large mammals was regularly eaten, a verydiversified diet is also indicated. Animal remainsinclude those of moose, bear, deer, dog, porcupine, rab-bit, bobcat, fox, beaver, muskrat, skunk, woodchuck,raccoon, squirrel, chipmunk, mouse, shrew, turtle andone or more varieties of bird. Riverine resources includefresh water mussel, shad, salmon, alewife, sucker, carp,perch, sunfish, catfish, bass and dace. Based on both theactual remains recovered and English documents of theperiod, com, beans and squash were stored in large quan-tity, but were supplemented by such other plant foods asblueberry, butternut, chokecherry, grape, elderberry, rasp-berry and/or blackberry, and staghorn sumac (Thomas1990).

By the fourteenth or fifteenth century, the subsistence baseof all communities living in Vermont probably combined amix of plants and animals acquired through traditionalhunting, fishing and gathering activities with therecently introduced domestic crops of com, beans andsquash (Figure 8). A generalized reconstruction of the sea-sonal exploitation practices of Indians living in the mid-dle and upper Connecticut River valley during the sev-


enteenth century follows.

In April, when the large spawning runs of alewives, shadand salmon began, families gathered by one of thelarge falls on the Connecticut River. The Abenaki lunarcalendar recorded in 1641 by John Pynchon inSpringfield, Massachusetts, suggests that at this time fishwas heavily consumed to the relative exclusion of othertypes of meat. By early to mid May, com fields were pre-pared and seeds planted. Fishing continued throughoutthis period, although some hunting of migrating water-fowl and some mammals occurred. By summer, familieswere centered around the family planting fields wherecom, beans and squash were tended. Throughout the sum-mer, a wider variety of fish could probably be taken in therivers, tributary streams and ponds. While crops werematuring in summer, wild plant foods, medicinal herbs andplant fibers were gathered.

By late August or September, horticultural produce wasedible. The collection, drying and storage of com, beans,squash, nuts and berries were undertaken during early fall.Once these activities were completed, a major deer hunt-ing period began. Although a number of the elderly andyoung children stayed behind in lowland hamlets, smallbands or single families established hunting lodges in theuplands where they remained isolated until late December.Here they hunted, dried or smoked deer and perhaps bearmeat for late winter consumption. From December untilmid-April, lowland settlements again reached their max-imum populations. Stored foods were relied upon heavilyand supplemented by local hunting, particularly formoose when heavy snows retarded their escape. Winter icefishing may have added variety to the diet. During at leastthe latter part of the Late Woodland period, then, com-munities were anchored to the lands adjacent to theirplanting fields in summer and winter, but also utilized bothriverine and upland sites as they focused on spring fish-ing runs or the fall hunt. With some variation, such asubsistence/settlement pattern probably characterized allAbenaki communities in the Connecticut valley betweenVernon and Newbury, Vermont (Thomas 1976).

Employing similar techniques, in conjunction with themajor fisheries and extensive wetlands along LakeChamplain. Indians in the Missisquoi, Lamoille, Winooskiand Otter Creek valleys of western Vermont probably usedthe uplands to the east much as communities in the middleConnecticut Valley used the uplands of eastern Vermontand western New Hampshire. It seems likely that theHoosic, Walloomsac and Batten Kill watersheds of south-

89


western Vermont were persistently used by Mahican orother non-Iroquoian communities whose horticulturalsettlements were located in the middle and upper Hudsonvalley.

Much remains to be learned about how different LateWoodland sites functioned, how various types of sites wereintegrated into larger settlement systems, and how suchsystems may have changed through time, particularly ascommunities shifted to an increasing use of horticulture.

Ideology

Few artifacts or features related to the spiritual beliefs ofthe Abenaki people of Vermont during the Late Woodlandperiod have been identified. Mudstone concretionswhich have been notched and/or etched, shell figurinesand other items may be tangible reminders that theAbenaki cosmology was substantially different than thatof the Western world (Thomas 1990). Given the con-tinent-wide distribution of sweat lodges for ritual purifica-tion at the time of European contact, sweat lodges are like-ly to be present in Vermont. One was apparently encoun-tered during excavations at an isolated site on a small ter-race along Otter Creek, but its age is unknown (Thomasand Bumsted 1980).

Elaborate burial ceremonialism seems to have become apractice of the past. During the sixteenth and seventeenthcenturies, and probably for at least several centuriesbefore, individuals were frequently buried in flexed posi-tions in shallow graves. The body was generally orientedon its side, with the head facing south and the face to theeast. Alternative orientations are also known, particularlyin the Connecticut Valley where individuals have report-edly been found in a sitting position. Burial inclusionswere rare prior to the arrival of Europeans and the incor-poration of metal and cloth items of English, French andDutch origin into Native American exchange systems.Graves of single individuals have been accidentallyencountered throughout the Champlain Lowland and alongthe Connecticut River; clusters of graves have been identi-fied in some areas which may represent family or commu-nity cemeteries (e.g., Bailey 1939). It has been suggestedthat burial practices actually shifted from rather randominterments at the beginning of the Late Woodland periodto a return to specific burial places and even ossuariestowards the end of this period, but this inference remainsspeculative (Heckenberger and Petersen 1988:24).

Late Woodland Period: References

Bailey, J. H.1937 A Rock Shelter at Fort Ticonderoga. Bulletin ofthe Champlain Valley Archaeological Society, Vol. 1(1):4-16.

1939 Archaeology in Vermont 1938. Archaeology inVermont [revised edition], edited by 1. C. Huden, pp. 3-6.Rutland, VT: Charles Tuttle, Co.

1940 A Stratified Rock Shelter in Vermont. Proceedingsof the Vermont Historical Society, Vol. 8 (1):3-30.

Chapdelaine, C.1989 The Mandeville Site and the Definition of a NewRegional Group Within the Saint Lawrence IroquoisWorld. Northeast Anthropological Association Abstracts,Annual Meeting, March 15-19.

1993 The Maritime Adaptation of the Saint LawrenceIroquoians. Man in the Northeast, Vol. 45:3-19.

Crisman, K. 1.1980 The Lower Missisquoi River Region: AnArchaeological Investigation. Honors thesis, Departmentof Anthropology, University of Vermont, Burlington.

Funk, R. E.1976 Recent Contributions to Hudson Valley Prehistory.New York State Museum, Memoir 22. Albany.

Heckenberger, M. and 1. B. Petersen1988 Archaeological Investigations at the SkitchewaugSite: A Multicomponent Stratified Site in Springfield,Windsor County, Vermont. Report prepared for theVermont Division for Historic Preservation,Archaeological Research Center, University of Maine atFarmington, Farmington, Maine.

McNeish, R. S.1952 Iroquois Pottery Types. National Museum ofCanada Bulletin 124. Ottawa.

Martijn, C. A.1990 The Iroquoian Presence in the Estuary and Gulf ofthe Saint Lawrence River Valley: A Reevaluation. Man inthe Northeast, Vol. 40:45-63.

90

Pendergast, 1. F.1990a Emerging St. Lawrence Iroquoian SettlementPatterns. Man in the Northeast, Vol. 40:17-30.

1990b Native Encounters with Europeans in the SixteenthCentury in the Region Now Known as Vermont. VermontHistory, Vol. 58(2):99-124.

Petersen, J. B.1990 Evidence of the Saint Lawrence Iroquoians inNorthern New England: Population Movement, Trade, orStylistic Borrowing? Man in the Northeast, Vol. 40:31-39.

Petersen, 1. B., M. 1. Heckenberger and P. A Thomas1988 Archaeological Investigations at the Pearl StreetPark Sites in Essex, Chittenden County, Vermont.Department of Anthropology, Report 95. University ofVermont, Burlington.

Petersen, 1. B. and D. Sanger1990 An Aboriginal Ceramic Sequence for Maine and theMaritime Provinces. In Pehistoric Archaeology in theMaritimes, edited by M. Deal. Council of MaritimePremiers, Fredericton.

Petersen, 1. B., 1. A Wolford, N. D. Hamilton, L. LaBarand M. Heckenberger1985 Archaeological Investigations in the ShelburnePond Locality, Chittenden County, Vermont. Annals ofCarnegie Museum, Vol. 54 (3):23-75.

Ritchie, W. A1969 The Archaeology of New York State [revised edi-tion]. Garden City: Natural History Press.

Snow, D.1980 The Archaeology of New England. New York:Academic Press.

Thomas, P. A1976 Contrastive Subsistence Strategies and Land Useas Factors for Understanding Indian-White Relations inNew England. Ethnohistory, Vol. 23 (1):1-18.

1990 In the Maelstrom of Change: The Indian Trade andCultural Process in the Middle Connecticut River Valley:1635-1665. New York: Garland Publishing.

Thomas, P. A and W. A. Bayreuther1980 Wells RS 0145(7): Phase I Archaeological

Assessment. Department of Anthropology, Report 25.


University of Vermont, Burlington.

Thomas, P. A and M. P. Bumsted1979 Burlington-Colchester M 5000(3) Connector:Phase II Intensive Archaeological Survey. Departmentof Anthropology, Report 6. University of Vermont,Burlington.

Thomas, P. A, G. Campoli and P. Doherty1979 An Initial Study of the Prehistory and HistoricArchaeological Sensitivity of the Bennington Area.Department of Anthropology, Report 22. University ofVermont, Burlington.

1983 Intensive Archaeological Evaluation for SelectedPrehistoric Sites within the Highgate FallsHydroelectric Project Area and PrehistoricArchaeological District rVT-FR-I03, 104, 134, 139, 140J.Department of Anthropology, Report 44. University ofVermont, Burlington.

Thomas, P. A. and R. S. Dillon1985 The Highgate Converter Station Site rVT-FR-161]. Department of Anthropology, Report 55.University of Vermont, Burlington.

Thomas, P. A, P. Doherty and E. 1.Warren1982 Cultural Resource Management Study, BallMountain Lake, Jamaica and Londonderry, Vermont.Department of Anthropology, Report 39. University ofVermont, Burlington.

Thomas, P. A and E. J. Warren1984 Reconnaissance Survey and Site Evaluation for VT-WD-36 Ball Mountain Lake, Jamaica, Vermont.Department of Anthropology, Report 51. University ofVermont, Burlington.

Wright, 1. V.1979 Quebec Prehistory. Archaeological Survey ofCanada, National Museum of Man, Ottawa.

91

Documents

Vermont Archaeology Comes of Age: A Perspective on Vermont ... · Introduction Throughout prehistory people had to adapt to the world around them. This is not to say that the environment