The University of Maine DigitalCommons@UMaine
Electronic Theses and Dissertations Fogler Library
12-1984
Pattern and Variation in Prehistoric Lithic Resource Exploitation in the Passamaquoddy Bay Region, Charlotte County, New Brunswick
Anita L. Crotts
Follow this and additional works at: https://digitalcommons.library.umaine.edu/etd
Part of the Archaeological Anthropology Commons
Recommended Citation Crotts, Anita L., "Pattern and Variation in Prehistoric Lithic Resource Exploitation in the Passamaquoddy Bay Region, Charlotte County, New Brunswick" (1984). Electronic Theses and Dissertations. 3218. https://digitalcommons.library.umaine.edu/etd/3218
This Open-Access Thesis is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of DigitalCommons@UMaine. For more information, please contact [email protected]. PATTERN AND VARIATION IN PREHISTORIC LITHIC
RESOURCE EXPLOITATION IN THE PASSAMAQUODDY BAY REGION,
CHARLOTTE COUNTY, NEW BRUNSWICK
By Anita L. Crotts
An Abstract of the Thesis Presented in Partial Fulfillment of the requirements for the Degree of Master of Science (in Quaternary Studies). December, 1984
This study examines some factors that affected choice and use of lithic resources among prehistoric peoples living within a restricted geographical area during a period of 2500 years. The raw materials of chipped stone tools from six shell middens, habitation sites in the
Passamaquoddy Bay region of southwestern New Brunswick, are identified. These include rocks that are atypical of the regional geology. Source areas of indigenous materials are located. Canoe transport made the native resources accessible to the inhabitants of the sites under study, and probably facilitated acquisition of non-native rocks.
The effect that distance had on lithic resource exploitation is determined by comparing the relative proportion of each rock type within and between site assemblages. A rock type is commonly present in greatest proportion in a site near its source. People also exploited more removed rock resources within the Passamaquoddy Bay area, but to a lesser extent. Rocks foreign to Passamaquoddy Bay account for nearly half of the artifacts in the majority of the sites. 2
Different rock characteristics are associated with artifacts of different forms (unifaces and bifaces). Finer-grained, homogeneously- structured rocks were chosen more often for unifaces than for bifaces.
This correlation applies both to indigenous and foreign materials.
Despite changes in uniface form, little change in rock type is demonstrated by unifaces from the Aceramic (3000-2000 B.P.) and
Ceramic (2000-500 B.P.) periods. This is a pattern unlike that seen in many sites of the Maine-Maritimes region. A change in material selection is evident for stylistically-different stemmed bifaces of the Aceramic and Ceramic periods. Paralleling a regional trend, more fine-grained lithologies are chosen for Ceramic period stemmed bifaces. No change in dependence on the Passamaquoddy Bay lithic resource base for unifaces or for bifaces during the 2500-year period of prehistoric occupation is evident. PATTERN AND VARIATION IN PREHISTORIC LITHIC
RESOURCE EXPLOITATION IN THE Ij’ASSAMAQUODDY BAY REGION,
CHARLOTTE COUNTY, NEW BRUNSWICK
By
Anita L. Crotts
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Master of Science
(in Quaternary Studies)
The Graduate School
University of Maine at Orono
December, 1984
Advisory Committee:
David Sanger, Professor of Anthropology and Quaternary Studies, Thesis Advisor Robson Bonnichsen, Associate Professor of Anthropology and Quaternary Studies Bradford A. Hall, Professor of Geological Sciences ii
ACKNOWLEDGEMENTS
The completion of this project was made possible by the
assistance of many organizations and individuals. Financial support was provided by National Sigma Xi, by the Maine Chapter of Sigma Xi,
and by a Graduate Grant-in-Aid from the University of Maine, Orono.
The Passamaquoddy Bay archaeological project was funded by the
following: The National Museums of Canada, The National Science
Foundation, the Province of New Brunswick, and the University of
Maine, Orono.
I am especially indebted to David Sanger, whose patience and
direction guided me through very difficult times, and to A. A.
Ruitenberg, Frank Howd, Bradford Hall, Robson Bonnichsen, and John
Stirling for their professional interest and input. Stephen
Bicknell’s draftsmanship and photographic skill are also greatly
appreciated. There have been many others who over the past several years have lent their support and furnished suggestions towards the
improvement of this work. Those to whom I am particularly grateful
are my good friend Diane Kopec, Margaret Chernosky, Chris Borstel, Jim
Steele, my aunt and uncle, Eva and Homer Brittain, and my brother and
sister, Arlin and Aletha Crotts.
This paper is dedicated to the memory of my parents, Adylade and Arlin Crotts. Of all the assistance I have been fortunate to receive, it is to their support and inspiration I am the most indebted. iii
TABLE OF CONTENTS
Page
LIST OF TABLES vi
LIST OF FIGURES viii
LIST OF PLATES ix
Chapter 1 INTRODUCTION 1
Chapter 2 THE GEOLOGICAL AND ARCHAEOLOGICAL FRAMEWORKS 6
LOCATION OF PASSAMAQUODDY BAY AND THE ENVIRONMENTAL SETTING 6
THE GEOLOGICAL FRAMEWORK 6
Glacial History 6
History of Bedrock Geological Investigations 7 in the Bay Region
Bedrock Geology 8
Geologic History 10
THE ARCHAEOLOGICAL FRAMEWORK 11
Prehistoric Peoples and the Geophysical Environment 11 of Passamaquoddy Bay
History of Archaeological Investigations in the 11 Bay Region
Cultural Reconstructions and Selection of Sites 15 for Study
Excavation of the Selected Sites 18
The Carson Site (BgDr5) 19
The Orr’s Point Site (BgDr7) 20
The Teacher’s Cove Site (BgDrll) 22
The Eidlitz Site (BgDs4) 24
The Minister’s Island Site (BgDslO) 25
The McAleenan Site (BhDrl) 27 iv
Page
Problems in Interpreting Shell Midden Stratigraphy 28
Selection of Tool Groups and Samples 31
DETERMINING AREAS OF LITHIC RAW MATERIAL EXPLOITATION 37
PASSAMAQUODDY BAY LITHIC MATERIALS 38
1. White Quartz 38
2. Grey Quartzite 41
3. Porphyritic Tuff/Rhyolite 42
4. Black Siltstone 44
5. Black Volcanic 46
LITHIC MATERIALS FOREIGN TO PASSAMAQUODDY BAY 48
1. Cryptocrystalline Quartz 48
2. Green and Red Mudstone 52
3. Green Volcanic 54
4. Ferro-manganese Metasedimentary Rock 56
5. White-spotted Metasedimentary Rock 57
Chapter 3 PATTERNS OF LITHIC MATERIAL SELECTION 60
INTRODUCTION AND PROCEDURE 60
THE EFFECTS OF DISTANCE ON RAW MATERIAL SELECTION 62
Exploitation of Nearby Lithic Resources 66
Exploitation of Distant PB Resources 67
Utilization of Imported Material 70
TOOL CLASS AND RAW MATERIAL SELECTION 71
Exploitation of PB Lithic Resources for the 72 Manufacture of Unifaces and Bifaces
Selection of Imported Materials for Unifaces and 72 Bifaces
Pattern Integrity Within the Biface Group 76 v
Page
Comparison of stemmed biface and thin 76 nonstemmed biface material selection
Biface fragments and material selection 78
Thick NSBs and material selection 78
Tool Class and Additional Raw Material Characteristics 78
TEMPORAL PATTERNS OF RAW MATERIAL SELECTION 84
The Temporal Framework 84
Traditions of Uniface and Biface Material Selection 92
COMPARISON OF THE SOURCES OF VARIATION IN THE RESOURCE 100 EXPLOITATION PATTERNS
Chapter 4 CONCLUSIONS 103
Factors That Affected Lithic Resource Selection and 103 Exploitation
Resource distance 103
Tool class 103
Resource distance and tool class 104
Time and tool class 104
Time and resource distance 105
Coastal Adaptation and Lithic Resource Exploitation 105
BIBLIOGRAPHY 108
APPENDIX 115
BIOGRAPHY 118 vi
LIST OF TABLES
Table Page
1 Selected sites and their cultural affiliations 17
2 Percentage distribution of PB and foreign lithic 63 materials identified in each site
3 Proportions of untyped lithic materials in each site 64
4 General locations of PB rocks 65
5 Sites located within each sector 65
6 Distribution of eastern rocks in eastern sites 67
7 Distribution of eastern rocks in western sites 68
8 Distribution of PB rocks among the sites 69
9 Distribution of foreign materials in PB sites 69
10 Unifaces and bifaces of identified materials: 71 site totals
11 Distribution of PB materials among unifaces from 73 each site
12 Distribution of PB materials among bifaces from 73 each site
13 Distribution of foreign materials among unifaces 74 from each site
14 Distribution of foreign materials among bifaces 74 from each site
15 Distribution of foreign materials in each tool group 76
16 Comparison of stemmed biface (SB) and thin nonstemmed 79 biface (NSB) material levels in each site
17 Biface fragments: material distribution 80
18 Thick bifaces: material distribution 81
19 Proportions of complete unifaces 10 g or > 10 g 89 in each site vii
Table Page
20 Temporal comparison of complete uniface materials in 94 BgDslO and BgDrll
21 Temporal comparison of stemmed biface materials in 94 BgDslO and BgDrll
22 Temporal distribution of PB and foreign materials in 96 unifaces
23 Temporal distribution of PB and foreign materials in 97 bifaces
24 Distribution of volcanic varieties in Aceramic and * 98 Ceramic stemmed bifaces
25 Stemmed bifaces: volcanic texture and temporal 99 association viii
LIST OF FIGURES
Figure Page
1 Location of Passamaquoddy Bay and selected sites 3
2 Bedrock geology of Passamaquoddy Bay 9
3 The Carson site 21
4 The Teacher’s Cove site 23
5 The Minister’s Island site 26
6 Temporal distribution of Passamaquoddy Bay sites 29
7 Source areas of exploited Passamaquoddy Bay lithics 40
8 Weight distributions of complete unifaces from each 91 site ix
LIST OF PLATES
Plate Page
1 Formed unifaces from the Carson site 33
2 White quartz 39
3 Grey quartzite 39
4 Porphyritic tuff/rhyolite 43
5 Black siltstone 43
6 Black volcanic 47
7 Red cryptocrystalline quartz 47
8 Yellow cryptocrystalline quartz 50
9 Grey cryptocrystalline quartz 50
10 Green mudstone 53
11 Red mudstone 53
12 Green volcanic 55
13 Ferro-manganese metasedimentary rock 55
14 Artifacts of white-spotted metasedimentary rock 58
15 Thin nonstemmed bifaces from the Carson site 77
16 Thick nonstemmed bifaces 82
17 Selected stemmed and nonstemmed bifaces 86 Chapter 1
INTRODUCTION
This study focuses on factors that affected prehistoric selection and use of lithic resources that provided raw material for making chipped stone tools. Now, as in the past,
Choices of usable resources, decisions as to their proportional use and time of utilization, and the demographic and. spatial arrangements chosen in order to accomplish the exploitation, all allot human time and energy and are visualized as structuring the subsistence and settlement patterns of a human group [Jochim 1976:4].
An inventory of present-day resources located near an archaeological site may not include those resources which were used by its inhabitants. Some exploited resources may not be recognized by the researcher, and others that seem important may not have been used.
Correlation of excavated organic materials with off-site resources is complicated by changing distributions, behavior, and content of biological communities.
Changes in the spatial relationships of sites and lithic resources, however, are less often produced by changes in the resources themselves. Culturally-removed lithic materials can be traced to their natural point of origin by a geologist. In addition, when the range of materials available within a contained geographical setting is known, rocks introduced from elsewhere can be identified in a collection. Prehistoric peoples’ dependence on different lithic resources will be expressed in the relative amounts of each rock in a site’s stone tool assemblage. 2
The Passamaquoddy Bay region of southwestern New Brunswick provides a setting for examining the spatial relationship of prehistoric settlements with lithic resources through time (Figure 1).
The bedrock geology of this 14 kilometer-wide extension of the Bay of
Fundy is characterized by multiple discrete units of different
lithologies outcropping along its northern perimeter. In the western
sector are eroding sandstone conglomerates containing clasts of various materials. To the east, the lithology changes to tuffaceous volcanic flows. Glacial drift covers the beaches.
Scattered along the coastline are eroding shell middens, habitation sites that were occupied during the 2000-500 B.P. period.
To judge from the sites1 locations near waterways, and from the
abundant remains of marine and terrestrial species, these people used watercraft, probably canoes, to exploit different biomes.
Historic accounts of aboriginal water travel in the Northeast describe the birchbark canoe’s large carrying capacity, maneuverability, and long range (Biard 1897:83, Erikson 1978:128,
Ganong 1899). Lithic resources that were too far away to be
effectively exploited via an overland route were easily reached in a day or less from any of the six sites chosen for study. Canoes would have also facilitated direct or indirect procurement of materials
foreign to Passamaquoddy Bay.
Canoes or dugouts were probably also used by the people who lived
in the Bay region 2000 to 3000 years ago. Frequently located beneath
the later shell middens, their shell-free deposits lack faunal
remains, bone tools, ceramics, and housepits. The only remaining
traits common to both cultures are chipped stone tools—nonstemmed 3
Figure 1. Location of Passamaquoddy Bay and selected sites 4 bifaces, and unifaces and stemmed bifaces that are stylistically
different. Nevertheless, the same lithic materials have been
identified in the assemblages from both time periods. The set of
utilized materials includes rocks that are similar to and others that
are atypical of rocks available in the Passamaquoddy Bay settlement
area.
The relative proportions of indigenous and imported goods in a
site’s material record is the evidence on which interpretations of
population mobility and intergroup contact have been based (e.g.
Bourque and Cox 1981, Gramly 1980, MacDonald 1968, Reher and Frison
1980). Since the mechanisms by which all such items were brought into
a site vanished long ago, one must take care when proposing models of
trade, migration, territoriality, etc., to include additional
economic, social or technological indicators as supportive evidence
(Abler 1977:148-149, Salmon 1978). In some studies, contact with
distant populations or areas with quarries has been discounted using
as evidence the absence of foreign materials (Gramly 1980, Lahti et
al. 1981). People who chose not to utilize distant lithic resources
could have participated in a vigorous exchange system of perishable
goods or nonmaterial commodities, such as society members.
No evidence exists for how the people of Passamaquoddy Bay
obtained lithic raw materials. In the stone artifact collections, however, is information with the potential to augment our knowledge of
prehistoric subsistence and settlement in the region. Previous
research has reconstructed a way of life that varied little among even
spatially distant and temporally distinct sites of the Ceramic period
(Bonnichsen and Sanger 1977, McCormick 1980, D. Sanger 1971). Did the 5 settlements also exploit the same lithic resources? Did people make the greatest use of resources near their homes or, with the ease of water transport, did they choose to make heavy use of distant sources of rock?
Because other aspects of the Ceramic and Aceramic periods’ ways of life seem so different, one would expect divergent patterns of lithic resource selection and use to emerge. If, on the other hand, continuity is indicated, then the products of other patterned behaviors that did not contribute to the material record, or have yet to be discovered, may have also transcended apparent breaks in the cultural history of Passamaquoddy Bay. 6
Chapter 2
THE GEOLOGICAL AND ARCHAEOLOGICAL FRAMEWORKS
LOCATION OF PASSAMAQUODDY BAY AND THE ENVIRONMENTAL SETTING
Passamaquoddy Bay is bounded on the north and east by the mainland of New Brunswick, by that of Maine to the west, and by the
West Isles to the south (Figure 1) . The relief here is low to moderate: the highest point is Chamcook Mountain, elevation 194 meters. Several streams and rivers cut through a mixed hardwood
conifer forest supporting populations of deer, bear, beaver, small
carnivores and rodents. Numerous species of fish and shellfish,
birds, seals, and porpoise are adapted to the Bay’s cold waters.
Additional details on biota and climate are presented by Thomas
(1983).
THE GEOLOGICAL FRAMEWORK
Glacial History
During the Wisconsin, there were two glacial advances: the older moved southeasterly, the younger east-southeasterly, to judge by the
direction of numerous roches moutonnees and striae (some overlapping)
throughout the area. Most of the region is covered with till and
outwash materials: one particularly heavy area extends from the
eastern shore of the St. Croix River southerly into the St. Andrews
peninsula and Minister’s Island (Alcock and MacKenzie 1960, Gadd 1973).
Outwash channel deposits also cover an area around the northern and western perimeter of Bocabec Bay, and follow the Digdeguash River to
the east. From Digdeguash Harbour to the Maguadavic, morainic 7 materials have been found combined with marine deposits. Similar mixed deposits, located 85 km to the east in Saint John Harbour, have been dated at 13,000 and 13,325 B.P. (Gadd 1973:20).
With deglaciation, a broad ice lobe that had extended to Deer
Island receded northward following the St. Croix (Gadd 1973:21).
Associated dates indicate the Bay was free of ice by 12,300 B.P. (Gadd
1973:22). Post-glacial isostatic rebound is indicated by raised beaches and waterfalls near the mouths of the Digdeguash and the
Maguadavic Rivers (Ruitenberg 1968:1).
Increasing tidal amplitude, now standing at 6 to 9 meters, has eroded the coastline, exposing bedrock reflective of that farther inland (Cumming 1967, Grant 1970).
History of Bedrock Geological Investigations in the Bay Region
The regional bedrock geology has been the subject of study for over a century. The first published investigation was by Abraham
Gesner for the New Brunswick Provincial Legislature in 1839. The first of five papers discussing the geology of New Brunswick, this report treated the geology of St. Stephen to Saint John in very general terms. In 1872, an exhaustive work detailing the geology of southern New Brunswick was published by L. W. Bailey and George F.
Matthew through the Geological Survey of Canada. Numerous papers and maps have since been published, notably bedrock maps of the St.
Stephen (Alcock and Mackenzie 1960) and St. George (Alcock and Perry
1960) quadrangles, Cumming’s report on the general geology of
Passamaquoddy Bay (1967), and Hay’s report (1967), which discusses the sedimentary and volcanic rocks between St. Andrews and St. George. A later publication by Ruitenberg (1968) deals with Passamaquoddy Bay in 8 its entirety. The latter two reports correlate the geology of
Passamaquoddy Bay with that of the Eastport, Maine area, also the subject of several maps and papers.
Bedrock Geology
The following discussion of the general geology of Passamaquoddy
Bay is based largely on the papers by Hay and Ruitenberg. Regions outside this thesis’ study area are included to aid in the understanding of the geological history of the area as a whole.
Within Passamaquoddy Bay itself, the oldest recognized rocks are
Silurian (Figure 2, Unit 1). These are located on Deer Island, the
Mascarene Peninsula, and along both the Maine and New Brunswick coasts. They consist of submarine mafic and felsic tuff and flows, slate and siltstone.
Lower Devonian porphyritic basaltic flows, rhyolitic tuff and interbedded clastic rocks are located along the northeastern
Passamaquoddy Bay coast and extend westerly to the St. Croix River
(Unit 2). These rocks are intruded to the north by the Devonian St.
George Batholith which consists of granite and gabbro (Unit 3).
The youngest lithologic unit of the area is the Upper Devonian
Perry Formation (Unit 4), composed of red conglomerate and sandstone with minor basalt flows and intrusions. This unit underlies most of the St. Andrews peninsula and the tips of the northern peninsulas. It is exposed near the Maguadavic River and on several islands. The conglomerate contains clasts of Silurian volcanics and sedimentary pebbles and boulders. The formation’s characteristic red color is derived from both its red clasts and from a hematite stain (Rhoades
1963). 9
k.l <$> \' Xy \ <’ 5 \. ‘ !2-l O. V .Y- \A \ • X ‘ • 2 . • s- \r\;4^0;• • •2’ ’ • . • • X. •
A <*<-*■ j3»» Granite, gabbro, -> 1' V *> < V> diorite > - X
Figure 2. Bedrock geology of Passamaquoddy Bay (after Ruitenberg 1968) 10
Geologic History
A submarine volcanic belt extended along the present-day coast of
New England into southern New Brunswick during the Silurian and Lower
Devonian (Gates 1967, Ruitenberg and Ludman 1978). Lower Silurian volcanics and sediments are submarine in origin (rocks of Unit 1).
Lower Devonian volcanism and terrestrial deposits (Unit 2) are associated with a closing marine basin.
Middle Devonian Acadian orogeny deformed the Lower Silurian rocks. The St. George Batholith emplacement followed (Unit 3).
Finally, terrestrial deposits of the Perry Formation (Unit 4) accumulated in an orogenic basin. Plant fossils place this unit in the Upper Devonian (Bailey and Matthew 1872:200). 11
THE ARCHAEOLOGICAL FRAMEWORK
Prehistoric Peoples and the Geophysical Environment of Passamaquoddy
Bay
In the Passamaquoddy Bay area, a wide selection of lithic materials formed and altered by various geological processes was available to the makers of stone tools. Although people have been settled along the coast for more than 3000 years, there are no indications of prehistoric rock extraction activities. Any alterations of outcrops would have been erased by hundreds of years of erosive wave action. In addition, no traces of cobble and rock fragment gathering would have been left on the beaches.
Lacking direct evidence of prehistoric manipulation of the geophysical environment, one must turn to the end products of the manufacturing process, stone tools. By treating it as a geological specimen, one can link an artifact to Passamaquoddy Bay’s coastal areas containing outcrops of similar materials, or distinguish it from the local lithology by its atypical material composition.
Stone tools have been found in numerous archaeological sites, most of which are located along the Bay’s northern perimeter (D.
Sanger 1971). Easily recognized by their eroding shell deposits, the sites have been the targets of many professional and amateur investigations for at least the last 100 years.
History of Archaeological Investigations in the Bay Region
In 1881, a publication of the United States National Museum listed and described a number of sites along the coast of Maine and eastern New Brunswick (Baird 1881). Baird’s report emphasized the 12 identification of faunal materials retrieved from these sites. A comprehensive investigation of the Bocabec site (BgDrl), located on the eastern shore of Bocabec Cove, was undertaken by members of the
Natural History Society of New Brunswick in 1883. The next year,
George F. Matthew, already recognized for his geological studies in southern New Brunswick, published the findings of the Bocabec project, which had combined zoological, botanical, and archaeological interests
(Matthew 1884). Matthew’s approach to gathering and interpreting multidisciplinary information illustrates the project’s advanced investigative framework. In 1899, W. F. Ganong published a monograph with the Royal Society of Canada in which prehistoric and historic sites important to the history of New Brunswick were located and described (Ganong 1899). He also discussed the distribution of the province’s historic tribes and mapped their extensive travel routes.
No further archaeological studies in the Bay area were pursued until the 1950s. In 1959, under the sponsorship of the R. S. Peabody
Foundation, Douglas Byers supervised a survey of sites in New
Brunswick, and excavated one on the western shore of Bocabec Cove -
Holt’s Point (BgDrlO). Unfortunately, only a minor paper has since been published describing a group of incised pebbles recovered from the site (Fowler 1966) .
Three sites were partially excavated in the 1960s by Richard
Pearson of the National Museum of Canada (Pearson 1970). Located in or near St. Andrews, these sites are shell middens on Minister’s
Island (BgDslO), Pagan Point (BgDsl) and Sand Point (BgDs6). The first radiocarbon dates for the area were published in this report. 13
In 1968, full-scale investigation of prehistoric sites in
Passamaquoddy Bay began. Funded by the Province of New Brunswick,
Mary E. J. Sanger excavated the Eidlitz Site (BgDs4) in St. Andrews
(M. Sanger 1968). Two years later, David Sanger, with funding from the National Museums of Canada and the Province of New Brunswick, directed a two-year project in which more than fifty sites in the Bay area were located and eight were systematically excavated (D. Sanger
1971).
Shell middens were regarded not as mere homogenous repositories of artifacts, but as products of many dimensions of human activities.
Hence, housepits, hearths, trenches and other features contributing to the construction of shell midden sites became important focal points of investigation. One paper resulted from this work - an overview of the Bay project including a discussion of its cultural-historical implications (D. Sanger 1971). Sanger also discussed prehistoric material discovered during Temple University’s 1972 excavations on St.
Croix Island (D. Sanger 1973).
Three master’s theses were generated by Sanger’s fieldwork. In
1972, Jacque Lavoie of the University of Montreal completed a study of artifacts retrieved during Pearson’s and Sanger’s work at the Sand
Point site (BgDs6) (Lavoie 1972). Based on attribute analysis of the chipped stone artifacts and on pottery characteristics, Lavoie placed the site at the beginning of the Woodland, or Ceramic period (2000-500 years ago in this area).
The Teacher’s Cove Site (BgDrll), partially excavated by David
Sanger in 1970, was completed in 1972 by Stephen Davis and became the subject of his M.A. thesis at Memorial University, Newfoundland (Davis 14
1978). By the time work at this site had begun, the excavators were
able to recognize housepits in shell midden stratigraphy. Davis
demonstrated that housepits were the foci of various activities,
evidenced by their high concentrations of stone artifacts and debitage, food bones, bone tools, and pottery. More important, two
components - one Aceramic (pre-2000 B.P.), the other Ceramic - were defined, based on artifact assemblages and their vertical placement in
the site. The respective lack and presence of pottery distinguish the
two cultural periods.
The third thesis, completed in 1980 by James McCormick of the
University of Maine at Orono, examined the effects of site depositional and taphonomic processes on faunal remains, and considered improved means for their collection, data management and analysis (McCormick 1980).
Since 1974, fieldwork in the Passamaquoddy Bay area has resulted in one paper, a report of findings from a field school led by Sanger at the Pagan Point site (BgDsl) in St. Andrews (D. Sanger 1975a).
Three additional papers have been published, based on previous fieldwork by Sanger. These include a publication describing housepits of the Ceramic period (D. Sanger 1976) and another in which intra- and intersite comparisons of faunal assemblages from two shell midden sites were made (Bonnichsen and Sanger 1977). The latter study has demonstrated through data banking and computer analysis that faunal material is not randomly distributed through time and space in these sites.
A 1981 publication by David Sanger was written in response to
Louis Brennan’s contention that deposition of shell and incorporated 15 artifacts did not occur simultaneously, but that artifacts were intrusive upon a pre-existing shell matrix (Brennan 1977). Using significant overlap in dates from shell midden and housepit charcoal
I from two Passamaquoddy Bay sites as evidence, Sanger demonstrated that shell deposition was contemporaneous with house occupation, a determination of particular importance to this research paper (D.
Sanger 1981b).
Artifacts recovered during Matthew’s excavation of the Phil’s
Beach (Bocabec) site (BgDrl) remained in storage until 1979 when they were resurrected for examination by Jennifer Bishop of the University of New Brunswick. Drawing on artifact stylistic similarities with those from dated sites in the region, she concluded that this site had been occupied during the Middle and Late Ceramic periods in the Bay
(2000-1000 B.P. and 1000-500 B.P., respectively) (Bishop 1980).
The government of New Brunswick has sponsored additional fieldwork in the Passamaquoddy Bay region by Christopher Turnbull of the Historical Resources Administration, Frederiction, and Stephen
Davis of Saint Mary’s University. Areas surveyed included Minister’s
Island, Deer Isle, and Indian Island in the Bay’s southern reaches
(Davis 1982, Davis and Christianson 1980).
Cultural Reconstructions and Selection of Sites for Study
From data collected after decades of shell midden exploration, one is left with the impression that the culture they represent varied little among the sites through the last 1500 years of prehistoric occupation. Bone tool assemblages—hafted and unhafted beaver incisors and antler tines, beaming tools, harpoons, needles and awls—and stone tools—unifaces, stemmed bifaces and nonstemmed 16 bifaces—changed little during this period. Tools were found scattered throughout the middens, but they were more often found concentrated in housepits—oval, semisubterranean excavations that had been filled with alternating layers of crushed shell, charcoal, sand and gravel—at the rear of most sites. Postmolds helped define the perimeter of some housepits, which measured approximately four meters by three meters. These features had been the foci of various activities, to judge by the abundant chipping detritus, hearths, and food bones contained in their fill (Bonnichsen and Sanger 1977:112,
Matthew 1884:16-17, McCormick 1980:98-99, D. Sanger 1976:11).
Analyses of faunal remains recovered from the sites* features and dump areas, composed primarily of soft-shelled clam (Mya arenaria) valves, indicate not only that these settlements were seasonal (late fall to early spring) occupations, but reflect a way of life dependent on marine, estuarine, and terrestrial species. Indeed, the same species are consistently represented in the sites’ faunal assemblages, but there is considerable intersite variation in the counts for each species. The variation may be explained by the sites’ occupations of different microenvironments (Bonnichsen and Sanger 1977:123, McCormick
1980:111-113). However, conditions observed today may not be the same as those present when the settlements were occupied. Not only has rising sea level obliterated large portions of most of these sites, it undoubtedly has affected the habitats of at least the marine and estuarine biotic communities. Climate fluctuations and man’s increasing influence on the Passamaquoddy Bay environment would not have held the ecological system constant to the present.
One cannot be certain that intersite variation in subsistence 17 patterns is a function of the sites1 proximities to different biological resources when the potential exists for resource modification through time. Fortunately, since lithic resources are more stable, they provide the means for determining whether nearby settlements were geared to their exploitation, and whether the decisions that affected their exploitation remained constant throughout the period of prehistoric occupation. In Passamaquoddy
Bay, occupation extends to at least 3000 B.P. The Aceramic
(3000?-2000 B.P.) components of the shell middens lack evidence for settlement and subsistence behaviors. Aside from a multiple burial, the only remaining data set is stone tools. Locked in these is information about exploitation patterns of lithic resources.
Logically, the contents from more than one site must be examined to observe any patterned spatial relationship between resource and settlement location; ideally, more than one site from each time period is needed for determining temporal variation in the spatial pattern
(Abler 1975:22, Struever 1971:11). In a settlement area where more than one lithic resource was exploited, it is important that sites have a geographic spacing as wide as that of the resource locations.
In Passamaquoddy Bay, these requirements are met in six sites
(Figure 1, Table 1).
TABLE 1. SELECTED SITES AND THEIR CULTURAL AFFILIATIONS.
SITE NUMBER NAME CULTURAL AFFILIATION BgDr5 Carson Ceramic period BgDr7 Orr’s Point Ceramic period BgDrll Teacher’s Cove Aceramic and Ceramic periods BgDs4 Eidlitz Ceramic period BgDslO Minister’s Is. Aceramic and Ceramic periods BhDrl McAleenan Ceramic period 18
Excavation of the Selected Sites
Only one of the six sites was excavated independently of David
Sanger’s supervision - the Eidlitz site. The rest were either wholly or partially excavated under his direction. During the 1969 field
season, emphasis was placed on sites in the eastern portion of the
Bay, and included Orr’s Point, McAleenan and the the Carson sites.
Sites in the western sector were excavated in 1970, including
Minister’s Island and Teacher’s Cove.
Excavation was concentrated on the seaward section of the first sites that were dug; later, the rear portion of a shell midden was recognized for its housepit potential. Consequently, in earlier excavations, a long trench close to and paralleling the shore was dug.
As recognition of housepits improved, more squares were placed in the shallower sections at the rear and to the sides of the sites.
All sites were laid out in a north-south and east-west grid system. Squares were named either according to their relationship with the main axes or were assigned letter designations. The northwest corner of each one-meter square excavation unit was tied in with a datum, the highest point on the site. The datum was in turn tied in with the high tide mark at each site. Davis’ work deviated slightly from this system by using three-meter square excavation units, but retained the one-meter squares for recording purposes
(Davis 1978) . The midden matrix in all of the sites was removed by trowelling and passed through J-inch mesh screens. Although floatation was tried, midden material did not lend itself to this technique (D. Sanger 1975b). When appropriate, samples of charcoal, shell, and bone were submitted for radiocarbon dating. 19
Each site was excavated in 10 cm levels, since natural stratigraphy was rarely well-defined. Artifacts were designated by the square and level in which they were found. If more precise provenience information was required, three-dimensional measurements were taken. Features such as housepits and hearths were dug as units independent of the 10 cm level system. When stratigraphy was particularly noteworthy, profiles were sketched and photographed.
Digging was terminated when bedrock or marine clay was encountered.
Field notes, drawings, and photographs were compiled in the laboratory and the artifacts were cataloged. The artifact collections were housed at the University of Maine until they were returned to the
National Museum of Man, Ottawa, in 1978.
The Carson Site (BgDr5)
One of three sites located in Digdeguash Harbour examined in this study, the Carson site has escaped destruction by natural forces and potholers’ activities. This anomaly can be attributed to the site’s fortunate situation on a highly-resistant volcanic outcrop, and within a thick tree covering. The trees not only inhibit erosion, but mask the highly visible shell. The volcanics that front the site do not afford a suitable environment for extensive clam flats today. Lower sea level in the past undoubtedly produced a different shore topography.
Baird assessed the sites of the Digdeguash River’s eastern coast as ’’not very productive (1881:296).’’ The Carson site is one that disproves his evaluation. Directed by David Sanger in 1969, the
98-square meter excavation was rich in artifacts and features 20
(Figure 3) (D. Sanger n.d.). Although not recognized during the fieldwork, one possible housepit has been located at the rear of the site by postexcavation analysis. Clues corresponded to characteristics of housepits identified during the 1970 field season: high concentration of artifacts, low percentage of shell in the fill, and location far from the shore.
Two chipping stations were found outside the pit, as well as several scattered and intact hearths. Charcoal, flakes, burnt shell, and broken animal bones were usually associated with the hearths.
Three charcoal samples taken from different hearths yielded the following dates: 420 ± 90 B.P. (SI-2186), 925 ± 80 B.P. (S-510), and
1120 ± 65 B.P. (SI-2187).
Most artifacts are typical of the Late Ceramic period in the Bay
(1000-500 B.P.): narrow corner- and side-notched points, thick cord wrapped stick pottery, small unifaces, nonstemmed bifaces and worked beaver teeth. Bone tools in a variety of forms including toggle harpoon heads, points, needles, and a comb were also taken from this site. Three stemmed bifaces found at the base of the midden are attributed to an Aceramic occupation (D. Sanger, personal communi cation 1983).
The Orr’s Point Site (BgDr7)
The remains of this rapidly eroding site rest on an exposure of
Perry conglomerate on the eastern shore of the Bocabec-Digdeguash peninsula. A steep grade fronts the site, precluding the formation of clam flats. A dry stream bed that cuts through the site may have once been a source of fresh water for its inhabitants. 21
Figure 3. The Carson site 22
Prior to David Sanger’s excavation in 1969, little was known
about the site’s history. Thirty-five square meters paralleling the
shore were removed, revealing two major feature areas with high
concentrations of artifacts (D. Sanger 1975b). Coupled with a change
in stratigraphy from the surrounding midden, the evidence suggested
the remains of two houses.
No radiocarbon dates were determined for the site, although
artifacts stylistically resemble those from the Carson and McAleenan
sites, tentatively placing Orr’s Point in the Late Ceramic period.
The Teacher’s Cove Site (BgDrll)
This site is located in a small cove just east of Creighton
Point, the eastern boundary of Bocabec Bay. Surrounded by forest and
swamp, the site occupies a clearing covered with thick grass,
indicative of recent cultivation, A storm beach and nearby volcanic
flows make up the shore topography.
In 1970, 63 square meters were excavated by David Sanger’s crew
in an area confined to the front portion of the site (Figure 4) (D.
Sanger 1975b). Artifacts typical of the Ceramic period were found, particularly in two housepits recognized by their distinctive
stratigraphy and central hearths. Two radiocarbon samples yielded the
following dates: 1170 ± 100 B.P. (S-608, charcoal) and 1635 ± 60 B.P.
(S-609, shell). Three additional pits containing similar fill and
artifacts were discovered in the rear portion of the site during the
1972 excavations. Conducted by Davis, the site excavation increased
to a total of 182 square meters (Davis 1978). 23 24
In the lower layers, Davis found artifacts suggestive of a different, earlier occupation. Large unifaces and contracting and straight stemmed bifaces were situated below the shell in a layer lacking pottery. This assemblage is comparable to those found in a few Aceramic, or Late Archaic sites in Maine, and is tentatively dated at 3000-2000 B.P.
The Eidlitz Site (BgDs4)
The Eidlitz site is located on the southern side of Joe’s Point, an extension of the St. Andrew’s peninsula. Underlain by weakly-cemented Perry conglomerate and sandstone, the site has undergone extensive erosion by the sea, and disruption from the installation of a golf course. It has also been plowed for many years.
A shell midden ”of some importance” on Joe’s Point, corresponding to the location of the Eidlitz site was mentioned by Ganong
(1899:223). From 1967-1969, 40 square meters were excavated, mostly from a trench paralleling the shore (M. Sanger 1968). Housepits were yet to be recognized in the stratigraphy, although a few hearths were encountered. A number of artifacts were recovered, including cores, stemmed bifaces, unifaces, nonstemmed bifaces, abrading stones, and flakes. Most stemmed bifaces had either wide or narrow corner notches.
Pottery, copper fragments, bone fragments and tools, beaver teeth, and assorted historic debris constitute the nonlithic artifact inventory. Varieties of pottery decoration include dentate, rocker dentate and cordwrapped stick patterns. Pottery sherds are thick and 25
tempered with crushed rock. Middle to Late Ceramic period cultural
affiliation is indicated by artifact stylistic attributes and by a date of 1290 ± 80 B.P. (GAK-1888) produced by a charcoal sample from a hearth.
The Minister’s Island Site (BgDslO)
Located just south of a causeway linking Minister’s Island with
the mainland, this site is one of the largest in the Bay area.
Although surrounded by crumbling Perry conglomerate, BgDslO has been
spared from more erosion by the resistant volcanic rock on which it
rests. The site has been plowed heavily during the years, which has
levelled any undulations in the surface topography modelled by the
prehistoric occupants. A small stream flows nearby. Clam flats
exposed on the causeway during low tide may have been exploited by the
early inhabitants.
The site was first mentioned by Ganong in 1899, but it was not until the 1960s that it was systematically tested. Excavations by
Pearson amounted to 75 square feet (approximately seven square meters) and yielded groundstone axes and adzes, large scrapers, bifaces, bone tools, and food bones (Pearson 1970). During the 1970
field season, David Sanger increased the excavated area by 85 square meters in units paralleling the beach (Figure 5) (D. Sanger 1975b).
Four housepits were discovered at Minister’s Island. Chipped
stone tools and detritus, bone tools, food bones, pottery, and
firecracked rocks were found within and around the pits. Two dates were determined: House 1 at 1060 ± 140 B.P. (GSC-1674, charcoal), and
House 2 at 580 ± 120 (GSC-1580, charcoal), although the latter sample 26
Figure 5. The Minister’s Island Site 27 may have incorporated intrusive material (D. Sanger 1975b). Two shell column samples yielded the following dates: 410 ± 130 B.P. (GSC-1542) and 650 ± 130 B.P. (GSC-1445).
One type of feature found only at Minister’s Island was a pit containing the badly-decomposed remains of at least seven individuals. Situated below the site’s shell layer, this burial belongs to an earlier cultural period. Associated artifacts are large, well-made bifaces, ground and chipped-and-ground celts.
Stylistically, these and other tools retrieved from the midden resemble artifacts of the Aceramic period in the Northeast. Charcoal overlying the pit gave a limiting date of 900 + 180 B.P. (GSC-1581).
A charcoal date of 2370 ± 80 B.P. (Y-1293) from Pearson’s nearby excavations is probably closer to the actual age of the burial
(Pearson 1970). Hence, two components are indicated at Minister’s
Island - one Aceramic, the other, Ceramic, with housepits, pottery, expanding stemmed bifaces, small unifaces and the midden itself.
The McAleenan Site (BhDrl)
The McAleenan site is located on the eastern side of Digdeguash
Harbour, about one kilometer below the river’s bridge crossing. It is situated on dark volcanics near a small stream and opposite foreshore flats which today are exploited by clammers.
A small shell midden, it was partially excavated by David Sanger
in 1969. Although only 17 square meters were examined, 387 artifacts were catalogued. These included cordwrapped stick decorated pottery, expanding stemmed bifaces and small unifaces of the Late Ceramic period. Two dates correspond with these findings, one from charcoal 28 at 680 ± 160 B.P. (GSC-1313), the other from shell at 450 ± 130 B.P.
(GSC-1292). No major features were encountered during the digging, but one area produced a large quantity of burned fish bone (D. Sanger
1975b).
Problems in Interpreting Shell Midden Stratigraphy
Four of the sites (Carson, Orr’s Point, Eidlitz, and McAleenan) have only one component of the Middle (2000-1000 years B.P.) and/or
Late (1000-500 years B.P.) Ceramic period. The two remaining sites
(Minister’s Island and Teacher’s Cove) are two-component sites, each with a Middle and/or Late Ceramic component and evidence for an extensive earlier Aceramic occupation (30007-2000 B.P.). Figure 6 illustrates the sites’ temporal position relative to other dated sites from Passamaquoddy Bay. The chronological subdivisions were established by David Sanger using radiocarbon dates and pottery sequences from the area (D. Sanger 1973:17-18, 1980:33-34).
In this study, the unit ’’component” is an artificial construct based on distinguishable associations of cultural remains that appear to be exclusive to only one time period in the history of a site. As generalizations about past human behaviors, they are useful in intra- and intersite comparisons. Evidence for the presence of differing sets of behaviors in discrete segments of the occupational history of a site does not necessarily indicate the presence of different populations with different cultures. A long-term, continuous occupation could have been responsible for an archaeological record punctuated, for instance, by changes in tool kit, food acquisition behaviors, or habitation. On the other hand, populations of different cultural affiliations may each have left behind archaeological 29
o
250
500
750
1000
1250
1500
1750
2000
2250-
O 2 < 2500- cr UJ Q <
2750—
3000
Figure 6. Temporal distribution of Passamaquoddy Bay sites 30 assemblages in a site that now appear as one unified record.
Furthermore, a shell midden is no exception to a problem that plagues interpretation of many other kinds of sites: one site composed of the materials of a single, long occupation may be indistinguishable from another produced by several short-term occupations. No doubt, there are numerous other possibilities that complicate interpretation of the cultural history of a shell midden site and hinder comparison with other sites.
Shell middens, in particular, are sites in which the results of both depositional and postdepositional processes interfere with the unravelling of their occupational histories. These sites were built up in such a way - most likely, by individual dumping episodes - that strata with broad horizontal integrity are virtually nonexistent. As
David Sanger has pointed out (1980, 1981a, 1981b), different sections of a shell midden can accumulate at different rates, depending on the proportion of shell to other deposited materials. Differential rates of deposition thus hinder temporal correlations within a site. After deposition, shells decaying at varying rates - depending on species, shell size, degree of weathering or firing - result in differences in shell consolidation throughout a midden. In sections of the site containing coarsely-broken or whole valves, downward movement of small artifacts is possible. This may explain why, for example, gunflints were found deep inside housepit fill at the Minister’s Island site, and nails were encountered in aboriginal artifact layers at the
Fernaid Point site in Maine (D. Sanger 1980). Conversely, artifacts attributable to the Aceramic components of both the Minister’s Island and Teacher’s Cove sites have been found in upper prehistoric levels 31
and plowzones. The limitations on stratigraphic correlations imposed
by depositional and postdepositional processes are summarized by
Sanger:
Until we know how all sections of the site were formed, we will not be in a position to accurately assess the relationship between artifacts and the shell strata over substantial stretches of (the) site [D. Sanger 1981b:40].
In Passamaquoddy Bay, there remains enough stratigraphic
separation in the two multicomponent sites to distinguish many, but
not all, Aceramic from Ceramic materials. This is facilitated simply
by the association of the shell matrix with the Ceramic component;
that located beneath the shell is usually of the earlier, apparently
nonshellfish-exploiting component. Whether the break in stratigraphic
continuity parallels a break in cultural continuity at these sites has
yet to be established. In addition, the criteria for discriminating
shell strata of the Middle Ceramic period from that of the Late
Ceramic period have not and may never be determined. Nevertheless,
there is no strong evidence for any significant alteration in a way of
life lasting 1500 years that produced an archaeological record
apparently marked only by minor stylistic differences in artifacts.
Selection of Tool Groups and Samples
In all probability, the prehistoric flintknappers of
Passamaquoddy Bay chose raw materials for their structural and behavioral characteristics, with manufacturing technique and tool
function in mind. Determination of the function of stone tools is a difficult course of study; no functional analysis has been performed
on those from the PB sites. However, in the Northeast, as in most 32 parts of the world, artifact classes that have had the most utility as temporal correlates are those based on formal attributes; these probably crosscut many undetermined functional classes. Tool groups common to both Aceramic and Ceramic components in Passamaquoddy Bay sites—formed unifaces and bifaces—have helped place the sites in the regional prehistory.
Formed unifaces are unifacially retouched flakes with a steep convex working edge (Plate 1). They are also known as "end" or
"thumbnail” scrapers. A biface is a flake that has been thinned and shaped through the removal of smaller flakes from its dorsal and ventral surfaces. Most bifaces in the PB collections have transverse and longitudinal cross sections that are biconvex. Several forms of bifaces are pictured in Plates 15-17.
Other chipped tools, such as flakes with light edge modification, are excluded from this study because they have had little utility in intersite comparisons in the Northeast. Various kinds of ground and chipped-and-ground tools, such as celts, are not common to both time periods here and are not useful for purposes of temporal raw material comparison.
Unmodified flakes provide limited information and are also excluded. Although relative proportions of nonlocal and local flake materials have been used in many studies (e.g. Abler 1975, 1977;
Osborne 1965) , there are problems inherent in flake counts (or weights or sizes) that limit their usage as indicators of the extent of material importation or of local resource dependence. For instance, when one compares the number of flakes removed during the production of a large biface to those removed from a rock of equal size while 33
uiniiii iiiiiiiiiii niiiitiii iimiii cm| 1 ) I 2I 1 3| 1 4] 1 BgDr - 5 ! 5] A . 1 . IQ . 1 . h
Plate 1. Formed unifaces from the Carson site. 34 making a smaller biface, of what significance is the difference in flake numbers or weights? Although two tools were manufactured, more flakes were generated during the production of the smaller than during that of the larger biface. Conversely, blocks of materials of unequal size may, under the right circumstances, yield equal numbers of flakes during tool manufacture.
Small flakes were lost through the screen during the
Passamaquoddy Bay excavations, but their presence as proof of tool trimming in any site and, hence, tool curation (especially of imported materials) is invalidated when one considers that very small flakes can be removed anytime during the manufacturing process.
There are additional complications with flake counts in a situation where water transport is indicated. Had little-trimmed blocks of rock or unmodified cobbles been carried via water, more material would have been available to inflate the frequencies of flakes derived from foreign localities. Locally-available materials may have also been moved by water or overland. Those transported by foot, presumably, would have first been trimmed at the source to reduce the weight of the load. Thus, a reversal in expected flake frequencies could result. Because we do not know how local and nonlocal materials were obtained, prepared, and transported, this study must focus only on the tools recovered from the sites.
All bifaces and formed unifaces recovered from each site were treated as a sample representative of the site. Many assumptions that cannot be verified in the data are behind this decision. The first is that these represent all such tools ever incorporated, even momentarily, into the cumulative assemblage of chipped stone tools. 35
The number of tools that were exported, lost, reworked, scavenged, or have since been washed away or taken by amateurs will never be known.
Most sampling strategies are based not only on the assumption of unchanged site size, but of homogeneous distribution of artifacts within a site. Neither requirement is consistent with the depositional and postdepositional histories of the Passamaquoddy Bay habitation sites. Even if they had once been nonrandomly located in each midden, artifact placements would have been disrupted by the churning effects of subsequent environmental and cultural post depositional processes. Nevertheless, some distributional patterns—such as tools concentrated in chipping areas—are still evident in the sites’ dump areas.
The distribution of stone tools, like faunal materials, is concentrated more in the housepits than in the surrounding midden.
This pattern was observed by Matthew at the Bocabec (BgDrl) excavation:
...the arrowmaker of Bocabec conducted his operation chiefly within doors...By far the best work in this line of art at Bocabec was found within hut bottom A, where the chipping of the lance and arrow-heads was performed beside the fire-place on stone or supports near the fire. The flakes resulting from the manufacture of these implements were very plentiful in this part of the hut bottom. Very few flakes were found outside the hut, and these mostly beside stones used for wedging the poles of the frame-work that supported the covering of the hut. In the kitchen-midden, flakes are quite rare [1884:16-17].
Numerous "arrow and lance-heads" and scrapers were also found in the housepits (Matthew 1884:17-18). Similar artifacts taken from two
Teacher’s Cove (BgDrll) houses (or 8% of the site excavation) account for about 40% of all tools recovered from the site. At Minister’s 36
Island (BgDslO), 50% of the tools came from the three best-defined housepits (15 of 85 square meters) or 18% of the excavated area.
McCormick, in his study of bone distribution patterns, discovered that, at least at BgDslO, the faunal species variety was considerably less in the housepits than in the midden (McCormick 1980:102-103). In two BgDslO housepits, as many varieties of lithic materials are present as are in the shell. Between the two pits, there is a large discrepancy in the relative percentage of each material. For instance, the dominant material in Housepit 2 is a variety of mudstone, comprising 48% of all of this feature’s tools. In Housepit
4, 45% of the tools are made of quartzite. The remaining tools in each pit were made of seven other rock varieties, but no variety is as abundant as the mudstone or the quartzite. A similar pattern, but with different rocks, is present in the stone tools from two Teacher’s
Cove housepits.
If manufacturing discards were concentrated in the housepits, and worn or broken tools were thrown into the shell dump, then a strong possibility for nonrandom distribution of raw materials exists.
Furthermore, beach erosion may have skewed the distribution of lithic types in favor of those concentrated in the housepits located at the rear of the sites. Adding artifacts from the remaining forward portion of the site increases the number of tools made of each variety. Together, all housepit and midden tools are probably the best sample now available to represent the site and its lithic raw materials before it was reduced by erosion. 37
DETERMINING AREAS OF LITHIC RAW MATERIAL EXPLOITATION
Two geologists (A. A. Ruitenberg of the New Brunswick Department of Natural Resources, Mineral Resources Branch, Sussex; and John
Stirling, then of the Department of Geology, St. Francis Xavier
University in Antigonish, Nova Scotia) were consulted with respect to this research. From an assortment of flakes, they identified several varieties of material that could have been obtained from nearby outcrops. This provided a basis for the initial rough sorting of the stone tools into material groups.
Color, mineralogy, texture, and structure were visual attributes used for deciding the rock categories. Increasing familiarity with the materials, along with additional input from the geologists, in particular, Dr. Frank Howd of the University of Maine at Orono, resulted in the identification of most of the artifact materials.
Low-power magnification with a Bausch and Lomb binocular microscope assisted in making difficult identifications, such as of weathered artifacts. The material groups were checked for internal consistency by Howd and Dr. Bradford Hall, also of the University of Maine.
A sample of each group was examined by Ruitenberg who identified materials not found in Passamaquoddy Bay. He also identified rock units of the Bay region from which other materials probably originated. Many of the utilized PB materials are particularly distinctive and readily distinguishable in handspecimen. Numerous artifacts still retain the characteristic weathered surfaces they had when they were once part of the local outcrops.
Geochemical analysis was not recommended since it probably would not have greatly assisted in making artifact and PB rock unit 38 correlations that could be established visually (Ruitenberg, personal communication 1981) . It was neither performed on materials that did not originate in the Passamaquoddy Bay region as it is beyond the scope of this paper to identify foreign source areas. Finally, damage to artifacts was prohibited, which precluded the use of thin-section and many chemical analyses.
PASSAMAQUODDY BAY LITHIC MATERIALS
1. White Quartz (Plate 2) 34 Specimens
Characteristics of quartz include a hardness of 7, conchoidal fracture and vitreous (glassy) luster. In this study the quartz that was exploited has features characteristic of eroded vein material: internal fracture, rounded form and dull, reddish-orange weathered surface. Numerous broken quartz cobbles have been recovered from some of the sites. Many quartz tools were fashioned out of jagged fragments with a minimum of shaping. Fracture is weakly conchoidal.
Color is milky to greyish white; a few samples are translucent with yellow, orange, and red tints.
White quartz is ubiquitous in the Perry Formation as clasts and is especially abundant on the St. Andrews peninsula (Figure 7).
Bailey and Matthew (1872:201) observed quartz as cobbles 6-8 inches in diameter at the Joe’s Point (adjacent to BgDs4) exposures; their observation was confirmed during this author’s field investigation.
Small quartz pebbles as well as cobbles as heavy as 3500 g were noted.
All clasts bore red surfaces characteristic of the Perry Formation.
In some areas, as at Joe’s Point, pebbles were easily removed by hand; at others, they had to be chipped out. Because the Perry conglomerate 39
Plate 2. White quartz.
Plate 3. Grey quartzite. 40
Figure 7. Source areas of exploited Passamaquoddy Bay lithics 41 is weakly resistant to weathering, many of the beaches are littered with its former clasts which could have been gathered. Flintknapping attempts with a hard hammer demonstrated the friable fracturing qualities of this material.
The Perry Formation exposed near the Minister’s Island site
(BgDslO) also contains quartz. In two exposures near the Orr’s Point site (BgDr7), however, no quartz was observed. A few small pebbles were seen on the beach fronting the conglomerate.
Although there are no extensive conglomerate exposures near the
Teacher’s Cove site (BgDrll), there is a high percentage of quartz pebbles along the beach, some of which were sufficiently large for toolmaking. No quartz was observed near the McAleenan (BhDrl) or
Carson (BgDr5) sites.
Matthew (1884:20-21) summarizes the availability of quartz for toolmaking purposes:
A third rock, or rather mineral, quartz, was one of which the men of Bocabec availed themselves to a large extent in the manufacture of stone weapons. As this mineral occurs abundantly in the pebbles of the drift and other surface deposits of the region, it is plentiful on the sea beaches, which abound with stones washed out of these surface deposits, and no special source of supply need be looked for.
2. Grey Quartzite (Plate 3) 119 Specimens
Quartzite is a metamorphic rock derived from quartz sandstone.
It is characterized by a granoblastic texture in which the individual grains are tightly interlocked; fractures pass through rather than around the grains. The collections’ grey quartzite has a massive structure, contains no relict bedding, and has a well-developed conchoidal fracture. A few specimens contain white quartz veinlets. Color consists of shades of medium to dark grey (Munsell 1971:10 YR 4-5/1, 4YR 5-6/1,
2.5Y 3-4/0). A few examples of reddish-grey (10 R 3/1-2) quartzite are present.
Like quartz, grey quartzite is present in the Perry conglomerate
(Figure 7). Bailey and Matthew (1872:201) note its presence in the outcrops at Johnson’s Cove, approximately 3.5 km northwest of St.
Andrews. During a field check by this author, it was observed in conglomerate exposures adjacent to the town. Pearson (1970:184), reporting on the high number of quartzite artifacts from three nearby sites—Minister’s Island (BgDslO), Pagan Point (BgDsl), and Sand Point
(BgDs6)—surmises that many were manufactured from beach pebbles. All of these sites are near Perry exposures. The reddish stained surfaces observed on quartzite artifacts are characteristic of the Perry’s clasts.
No quartzite was observed in the Perry outcrops bordering Bocabec
Cove and Digdeguash Harbour. Exposures at the mouth of the Maguadavic
River were not examined. If quartzite is present here, it is not significantly closer to the easternmost sites than are the western PB exposures of the Perry conglomerate.
3. Porphyritic Tuff/Rhyolite (Plate 4) 44 Specimens
This group consists of extrusive volcanics with a porphyritic-aphanitic texture in which phenocrysts—crystals that can be seen with the naked eye—are set in an aphanitic matrix of crystals determinable only with a microscope. Although the groundmass may be one of several colors—black, green (Munsell 1971:5YR 4/1, 5GY 4/1), and brown (5YR 3/2-4, 5YR 2.5/1-2)—all specimens in this group 43
Plate 4. Porphyritic tuff/rhyolite.
Plate 5. Black siltstone. 44 contain pink (10R 4-6/8) feldspar phenocrysts with and without pyroclastic inclusions of the same color. The presence of these
inclusions is the group’s distinguishing feature. Some phenocrysts are elongate and aligned in apparent direction of flow, others are randomly scattered. Phenocryst size varies from less than one mm to slightly more than two mm. Small white feldspar and quartz inclusions are also present in lesser amounts, and a few scattered spots of magnetite are visible. Fracture is conchoidal.
There is little question that this material was derived from the extensive volcanic flows exposed along the northeastern shore of the
Bay (Figure 7) . Reddish-brown volcanics similar to but somewhat coarser than those in the collection are present in the vicinities of the Carson (BgDr5) and Orr’s Point (BgDr7) sites. The flow near Orr’s
Point has broken up into large blocks and debris, whereas the exposure near the Carson site remains smooth and unbroken. Similar conditions may have affected procurement techniques at other outcrops.
Green groundmass volcanics resembling those in the collections are exposed near the Teacher’s Cove site (BgDrll). Although no black volcanics with pink inclusions were observed in the field, it is assumed that they were obtained from the same unit as volcanics with a brown or green matrix (Ruitenberg, personal communication 1979).
4. Black Siltstone (Plate 5) 123 Specimens
Siltstone is a clastic sedimentary rock, an aggregate of small rock and mineral fragments—commonly quartz, mica, and clay minerals—in the 1/16 to 1/256 mm size range. The siltstones examined in this study are black or dark grey and are very fine-grained. There is no evidence of foliation, and conchoidal fracture is good. A few 45
specimens have white quartz veinlets.
Evidence for the probable location of this rock’s source was
derived principally from a literature search, in particular, from
Matthew’s (1884) discussion of the Bocabec site (BgDrl) excavation.
Many of the recovered chipped tools were made of ’’petrosilex,” an
obsolete term referring to material he defined as a "fine-grained,
siliceous, sedimentary rock, baked or hardened" and one that "splits
with a deep conchoidal fracture." Petrosilex occurred in two colors,
dark brown and black (Matthew 1884:20).
Artifacts excavated by David Sanger at BgDrl in 1969 were
examined; by far, the majority of the artifacts were manufactured from
a black siltstone fitting the description of Matthew’s "petrosilex."
This material was also identified in the artifact assemblages from
other sites that are under study.
Matthew (1884:20) recorded the source area for the petrosilex:
The source of supply for the material from which the greater part of the stone weapons and implements of Bocabec have been fabricated is not far distant from the site of the village. The whole northern and eastern side of Passamaquoddy Bay is bordered by trap rocks and sedimentary rocks...This part of the terrain which, in the most southerly outcrops in Charlotte County, consists of hardened silicious shales, is represented at Bocabec by a fine-grained petrosilex, exposed in Digdeguash Basin and probably also on Bocabec River...It is this material which the men of Bocabec found most advantageous for the fabrication of lance and arrowheads.
The Digdeguash Basin locality corresponds to Hay’s (1967) description of a narrow exposure in the Basin’s neck (Figure 7). An
Upper Silurian and/or Lower Devonian unit, it consists of "siltstone,
sandstone, and silty argillite (present) in thin interbeds within the 46 volcanic sequence. These interbeds commonly consist of thick-bedded,
finely laminated, dark grey siltstone” (Hay 1967:6). Many of the
siltstone archaeological specimens retain remnants of a flat, weathered, natural surface. A field check by this author confirmed
the presence of this material throughout the Digdeguash Basin area.
5. Black Volcanic (Plate 6) 59 specimens
This is a porphyritic-aphanitic extrusive volcanic with small white feldspar phenocrysts in a very fine black matrix. Phenocryst
size range from 0.4 to 2.0 mm; the majority of the specimens contain phenocrysts at the lower end of the size range. Conchoidal fracture
is very pronounced.
The source location of this rock is uncertain, although it probably is of local origin. According to Ruitenberg, it may be from
the volcanic flows from which the Group 3 rocks were derived, but he has not observed it in the field (Ruitenberg, personal communication
1977). Stirling also suggests that it is of local origin, but does not know its precise location (Stirling, personal communication 1977).
The material is very similar to the black volcanic rock previously
described, but lacks its pink inclusions.
The McAleenan site contains the highest counts of black volcanic
tools and an abundance of its detritus. Based on the material distribution patterns observed in other PB sites near lithic sources
(Chapter 3) , this author suspects that the source of the black volcanics is near the McAleenan site (Figure 7). Dark volcanic flows outcrop nearby in the Digdeguash Harbour-Digdeguash Basin area which furthers this impression of close source proximity. 47
Plate 6. Black volcanic.
Plate 7. Red cryptocrystalline quartz. 48
LITHIC MATERIALS FOREIGN TO PASSAMAQUODDY BAY
The following is a description of materials foreign to the geology of the Passamaquoddy Bay region that have been identified in the collections. While the nearest known locations containing similar rocks are discussed, it is not implied that they were the source areas for these materials.
1. Cryptocrystalline Quartz 124 Specimens
Cryptocrystalline quartz is composed of grains too small to be identified at magnification less than 50X. Composed primarily of
SiO^, it is formed either by silica deposition in a marine environment, or by total or partial replacement of a pre-existing rock by silica from groundwater or hydrothermal solution. As a result, cryptocrystalline quartz has been classified both as a mineral and as a sedimentary rock. Several varieties based on differences in color, texture, and mode of formation, are divided into two broad groups—fibrous and granular. The fibrous varieties, also known as chalcedony, contain submicroscopic bubbles of water. Granular cryptocrystalline quartz includes flint, chert, and jasper. All types of cryptocrystalline quartz have a hardness of about 6.5, a conchoidal fracture, a lack of cleavage, and a dull to waxy luster.
Cryptocrystalline quartzes observed in the Passamaquoddy Bay collections have a smooth texture.
Most of the cryptocrystalline quartzes examined were placed into three color groups described below. In the analysis (Chapter 3), however, they are combined and treated as one group. 49
Red cryptocrystalline quartz (Plate 7) 49 Specimens
Color: pale to dark red (Munsell 1971:10R 3-6/6
and 1 or 5-6/2). Tan, black and/or
white mottling; some black and blue
black banding present. A few specimens
are homogeneous in color.
Transparency: translucent to nearly opaque.
Luster: waxy to dull.
Yellow-brown cryptocrystalline quartz (Plate 8) 20 Specimens
Color: tan to light brown (Munsell 1971:10YR
4/2-4, 5/3). Little mottling, mostly
homogeneous in color.
Transparency: translucent.
Luster: waxy to dull.
Comments: pebble cortex present on some specimens.
Grey and grey-green cryptocrystalline quartz (Plate 9) 39 Specimens
Color: dark greenish-grey (Munsell 1971:5GY 4-5/1, 5G
4/1). Several specimens are homogeneous in color,
others have broken black banding.
Transparency: opaque.
Luster: dull.
Comments: pebble cortex present on some specimens.
Although cryptocrystalline quartz was greatly utilized, evidenced by its abundance in the artifact assemblages, it was not obtained from within the Bay area. No geologist consulted knew of a possible local source. Additionally, no cryptocrystalline quartzes were observed in the field or mentioned in geological reports for the area. The 50
Plate 8. Yellow cryptocrystalline quartz.
Plate 9. Grey cryptocrystalline quartz. 51 nearest exposure is about 40 km southeast, on Grand Manan Island
(Matthew 1884:19,20; Ruitenberg and Stirling, personal communications
1976). Jasper, agate and chalcedony are found along the island’s western shore. A few fragments were discovered during a brief field
examination along the north shore.
Similar materials can be found along the coast of Nova Scotia on
the Bay of Fundy (Sabina 1972). The coastal locality closest to
Passamaquoddy Bay is about 85 km to the east at the mouth of the Saint
John River, and contains agates, chalcedonies, carnelians and jaspers
(Sabina 1972). The area is recognized as a possible source for most of the utilized cryptocrystalline quartzes in the Passamaquoddy Bay
collections by Matthew (1884:20) and Ruitenberg (personal
communication 1976).
The southern shores of Washademoac Lake, located 50 to 60 km upstream from Saint John, were quarried by prehistoric stoneworkers.
The variety of material here is described in Abraham Gesner’s report:
On the south-east side of a small cove (Belyea’s Cove), the shore is strewed to the distance of half a mile with loose masses of hornstone, jasper, Egyptian jasper, chalcedony and quartz. The jasper is chiefly of a red colour and passes into a milky chalcedony, being arranged in spots and clouds, and shaded with smoky imitative figures. Associated with the jasper is the variety called Egyptian jasper, which is distinguished from the other by peculiar zones, circles and clouds of different colours. With these a few small pieces of carnelian were found...[Gesner 1839:60].
Matthew explored the southern lake shore, and described another
source as ’’strewed with numerous fragments of chalcedony, carnelian, agate and jasper’’ originating from thin seams along the banks (Matthew
1900:62-63). He also noted the high proportion of worked 52
cryptocrystalline quartz in several nearby prehistoric sites, which
suggested quarrying of these exposures. This area is recognized as
the nearest verified aboriginal quarry to Passamaquoddy Bay by the
National Museum of Man’s Archaeological Survey of Canada Lithic Source
Identification Programme. Headed by Sterling Presley, this is a
library of lithic samples from identified sources in Canada and the
United States.
The grey and grey-green specimens with black banding are visually
similar to material outcropping near Munsungun Lake, but it has not been determined if they originated from the same area.
2. Green and Red Mudstone (Plates 10 and 11) 120 and 23 Specimens
Mudstone is very similar to siltstone, being a clastic sedimentary rock, but is composed of rock particles less than 1/256 mm in size. Texture ranges from very fine-grained to nearly cryptocrystalline: both varieties have undergone partial silicificaton. Unlike the collections’ siltstones, there are fewer mudstones with a massive structure than with a slatey cleavage, which allows them to be split into parallel-sided fragments. Conchoidal fracture is stronger in specimens whose structure is less laminated.
Color includes green to grey-green (Munsell 1971:5GY 4-5/1, 5G 4/1) and dusky red (2.5YR 3-4/2, 10R 3-4/3). A few specimens of each color have black banding. Remnant cobble surfaces are still visible on several artifacts made of these materials.
Neither mudstone has been observed by Ruitenberg or Stirling in the Passamaquoddy Bay area. Both rocks are Ordovician in age and do not appear to be part of the local geology. They are also atypical of the nearest Ordovician unit, the Cookson Formation, which is 18 or 53
Plate 10. Green mudstone.
Plate 11. Red mudstone 54
more kilometers from Passamaquoddy Bay (Ruitenberg, personal
communication 1977). There are no other Ordovician units within a 40
km radius of Passamaquoddy Bay (Osberg et al. 1984, Potter et al.
1968).
Red mudstones are associated with Ordovician deposits in an
extensive mountainous belt ranging from north central Maine eastward
into northern New Brunswick (Bradford Hall, personal communication
1983). They are of particular interest in northern Maine where
Ordovician outcroppings near Munsungun Lake have been quarried from at
least the time of the Paleoindians in the Northeast (Bonnichsen
1977:13). Many of the surrounding workshop sites date to the Ceramic
period; it may be more than just coincidence that artifacts from the
Passamaquoddy Bay sites were manufactured from material remarkably
similar to that at Munsungun Lake. Green silicified mudstones were
also exploited here, but are unlike those represented in the
Passamaquoddy Bay tool assemblages.
3. Green Volcanic (Plate 12) 37 Specimens
This is a porphyritic-aphanitic extrusive volcanic rock with
small white feldspar phenocrysts (usually less than one mm long) set
in a grey-green (Munsell 1971:5G 4-5/1, 5GY 4/1) matrix. Lesser
amounts of quartz and mafic minerals are present. Fracture is
conchoidal.
Similar volcanics have not been reported as outcropping locally.
Neither Ruitenberg nor Stirling have encountered it during their
fieldwork; Ruitenberg observed that its lithology is foreign to what
is found in the Bay region (Ruitenberg, personal communication 1977).
However, archaeological specimens manufactured of similar material are 55
Plate 12. Green volcanic.
Plate 13. Ferro-manganese metasedimentary rock. Sample is from a site near Woodstock on the Saint John River, New Brunswick. 56 common in sites throughout Maine, where it is referred to as
’’felsite." One well-known source is Mt. Kineo on Moosehead Lake in west central Maine, where felsite was obtained from a talus slope by prehistoric Indians (McGuire 1908; Willoughby 1901). Unfortunately,
this material is indistinguishable from outcroppings in localities apart from Mt. Kineo (Bonnichsen 1977:12-13). Furthermore, it is
impossible to restrict the sources of felsite only to isolated outcroppings since glacial actions have overrun these areas, making
this material ubiquitous throughout much of the state. Nevertheless,
this rock was not obtained from the Passamaquoddy Bay area.
4. Ferro-manganese Metasedimentary Rock (Plate 13) 6 Specimens
This is a fine-grained silicious red (Munsell 1971:5R 2.5-3/4, 5R
2.5-3/2) sedimentary rock with black patches of magnetite intermeshed with patches of hematite. A few scattered mineral inclusions of quartz and white feldspar are present. Fracture is conchoidal.
Very few specimens manufactured from this material were
identified in the collections. These artifacts were examined by
Ruitenberg, who suggested that their material was derived not from the
Bay area, but from somewhere in Maine, or New Brunswick west of
Fredericton. It is recognized by Frank Howd of U.M.O. as similar to
that occurring in eastern Aroostook County, Maine. Louis Pavlides
(1962) identifies three specific outcrop areas in Maine, part of a discontinuous belt, 1) in the Houlton vicinity, 2) in a locality west of Caribou and Presque Isle, and 3) in the Maple and Hovey Mountains, west of Bridgewater. It is not surprising that Archaic sites excavated by David Sanger in the nearby Woodstock, N.B. area contained numerous artifacts and abundant debitage of this kind of rock. The 57
outcrops are approximately 130-180 air kilometers from Passamaquoddy
Bay.
5. White-spotted Metasedimentary Rock (Plate 14) 6 Specimens
Because it was not decided to designate artifacts of this material as a separate group until after the collections had been
returned to the National Museum of Man, its description was not
afforded the detail that has been given the previous lithic material
groups. Instead, this description is based on work by Paul LaPierre,
a former student in geology at the University of Maine, Orono, who
examined the artifacts from the Teacher’s Cove site (BgDrl 1) and
identified their raw materials in 1974. In hand-specimen, it is
described as a chert of ’’medium green groundmass peppered with cream
colored spots with faded indistinct borders.” It is aphanitic in
texture, and in thin-section "the spots are indistinguishable from the
groundmass” (LaPierre 1974:6). The spots’ boundaries appear to be
partially dissolved, possibly as a result of silicification.
In 1952, a sample of this material was examined by a
Massachusetts geologist, Rose Moffett, who determined that it had
"been subject to complete modification and recrystalizaton" but could
not determine if it was igneous or sedimentary in origin (Moffett
1952). The sample had been submitted by Isaac W. Kingsbury who had
obtained it from "a felsite outcrop at the west end of Hinkley Point,
Dennysville, Maine" in Cobscook Bay during his and Wendell S.
Hadlock’s archaeological investigations in the late 1940s (Kingsbury
and Hadlock 1951:25). Archaeological specimens of this material were
observed in the Moose Island (near Eastport, Me.) and Michner Point
(near Perry, Me.) sites during their survey and by Gramly in other 58
Plate 14. Artifacts of white—spotted metasedimentary rock. 59
Cobscook Bay sites (personal communication 1980). A stemmed biface
and "numerous waste flakes" in the Teacher’s Cove collection were also
identified by Douglas Byers (Davis 1978:29).
None of these six materials were observed in the glacial till and
outwash that has eroded onto the beaches. Special attention was
afforded the St. Andrews peninsula where there are heavy glacial
deposits. A section of the shoreline more than one kilometer long was walked several times in a search for materials similar to those which had been used prehistorically* but only quartz and quartzite were
found. Surveys of other areas with glacial deposits in Passamaquoddy
Bay yielded no materials resembling those in the six nonPB rock
groups. 60
Chapter 3
PATTERNS OF LITHIC MATERIAL SELECTION
INTRODUCTION AND PROCEDURE
The investigative framework for research of prehistoric lithic resource exploitation patterns in the Passamaquoddy Bay region is defined by six prehistoric sites, two cultural periods, and two classes of chipped stone tools—unifaces and bifaces—whose members were manufactured of Passamaquoddy Bay and foreign lithic raw materials. The influence that distance, tool class, and changing material requirements had on resource selection and use can be elicited from patterns formed by the relative proportions of different rocks represented in the sites* artifact collections. Different patterns are best observed" in data sets arranged to answer the following questions:
1. What effects did differing distances to resource areas have on
lithic selection?
2. Were different kinds of tools made of different rocks? Do the
rocks in one tool group have physical characteristic, such as
natural form, size, and texture, not common to materials in
another tool group?
3. Do the same resource distance/selection patterns seen in #1 exist
within each tool group?
4. Were stylistic changes within each tool group accompanied by
changing rock selection? Could different raw material
requirements have been a factor in selection change?
5. Did reliance on local resources vary through time? 61
Together, the answers to these questions will increase our understanding of prehistoric peoples’ utilization of one environmental resource in the Passamaquoddy Bay region.
Intrasite and intersite comparisons of material quantities are based on raw scores and frequency data instead of on statistical correlations. The entire sample is sufficiently large to be subjected to statistical testing, but it is of limited statistical utility when distributed among the six sites and 11 rock categories. Probability statements would not be particularly meaningful since the size of the statistical error would be very large relative to the size of each subsample. In addition, from the perspective used here, the existence of a rock type in a site, even if in very small quantities, is in itself meaningful, as is its absence. Either condition may be disregarded in a statistical procedure. Consequently, relative percentages of materials are used for comparative purposes, both within and between assemblages.
Given the size variation among the six assemblages, intersite variation in relative percentages of a rock type is subject to broad interpretation. For instance, in one PB site containing 213 chipped stone artifacts, the relative percentage value for one material observation is 0.5; in another assemblage with 32 members, the value is 3.1. It is not implied that a material was chosen six times more frequently in the second site; rather, it was rarely chosen in either site. When interpreting any pattern of material selection, the reader should remember to consider the sample size and count behind each relative value. 62
The distribution of identified lithic materials among the chipped
stone artifacts from the six Passamaquoddy Bay sites is presented in
Table 2. Eighty-five percent (695) of the 813 tools examined were of the materials listed, with the remainder distributed among the untyped rock groups listed in Table 3. ’’Unweathered" materials are those easily determined to be not of the rock types described in Chapter 2.
There are a number of varieties in the untyped, unweathered group, but for no variety were more than four specimens identified.
Some archaeological specimens of sedimentary, metasedimentary, and volcanic rocks were so altered by natural processes that they could not be placed in any of the identified material groups.
Additional rocks were weathered such that it was impossible to determine if they were sedimentary or igneous in origin. They were placed in the ’’Unknown" group. It is doubtful that any quartz or quartzite artifacts were placed in any weathered group since they probably would not have weathered beyond the point of recognition.
All archaeological specimens in Table 3 were excluded from further analysis. Their absence does not alter the relative percentage pattern of lithic selection presented in Table 2. The greatest increase in the upper limit for any identified rock in Table
2 would be of only 11% in BgDs4 if all unidentified rocks were found to be the same material—an unlikely possibility.
THE EFFECTS OF DISTANCE ON RAW MATERIAL SELECTION
The location of most prehistorically-exploited PB rocks that have been identified are in one of two nonoverlapping sectors along the northern perimeter’s area of settlement (Table 4). White quartz is 63 1 1 | | | | | | |
z^ Z-*s z**s \O r- CO (0) (1)
(5) (3) x-z m ^-Z (9) 1—I ^z (19) p v—z (19) (25) P (107)
00
PQ o 00 2.8 0.0 4.7 0.9 8.4 m kD CM 17.8 17.8
r—4 23.4 100.1 SITE.
1 | 1 1 1 1 H Z~^ z-\ z*-s Z-S z*s Z~x /—s Z->1 Z*S z—\ z^s z**s ^z M-Z m Pl m_z S-Z iz P H 42 PQ
1
(I) Z—s Z~X m CN (3) m X-Z (11) Q
M) 8*0
PQ CM r-
i—4 9.2 2.5 1 1 1 Z~X Z"X Z"x Z~X Z-X Z*x XV= CM i—1 CO o \D 1—1 X-Z x^z xz x-z x-z X^z O rd PQ co \D CT> o 00 B'S • • • • • co r—4 o CTi Z“x z-x =^= s£> CM T—4 CM r—4 r- x_z X-Z X_Z x_z x—Z 1—4 x-z Q MJ co PQ o co p- p- B'S • • • • • M Z~x Z~X z*x Z-X Z“X =rt= CT' \O CM in CM O X^Z x_z i—4 X^z x-Z co i—1 x^Z x^ co •a r~ m 00 o o PQ B'S • • • • • CO \O o CM co r—1 Z~X Z~x Z“X Z-x Z“X Z~x =T$= H' o co m in co 4-1 0) e • • o nd r“"4 a) o co > T3 a) cn s-i nd nd a> a> CD o 42 H £ <1 co o ■u 4-> Fl H £ CO CO 4-1 Fl CD cd Fl —> & 23 H 65 present as drift in both areas, but it is also found as inclusions in the Perry conglomerate. TABLE 4. GENERAL LOCATIONS OF PB ROCKS. WESTERN SECTOR EASTERN SECTOR Grey quartzite Porphyritic tuff/rhyolite Quartz Black siltstone Black volcanic Quartz The sites under study that are located either within or reasonably close to (maximum distance approximately 4 km) units of these materials are listed in Table 5. Teacher’s Cove (BgDrll) is located such that it may have been equidistant from sources of western and some eastern rocks, but it is treated as an eastern site due to its closer proximity to most sites and rocks of that sector (Figure 7). TABLE 5. SITES LOCATED WITHIN EACH SECTOR. WESTERN SECTOR EASTERN SECTOR Eidlitz (BgDs4) Teacher’s Cove (BgDrll) Minister’s Island (BgDslO) Orr’s Point (BgDr7) Carson (BgDr5) McAleenan (BhDrl) Canoeists from any of the six settlements could have reached and returned from most of the PB lithic resource areas within one day. The maximum site-resource area distance following the coast is approximately 16 km, but other site-source distances are much shorter. From a geographical perspective, the settlement area is so restricted 66 in size that the distance to a particular foreign resource area may have differed little among the sites. Had these people made equal use of all materials available to them, there would be little variation in the rock levels among the sites’ artifact assemblages. As illustrated in Table 2, however, the distribution of materials within and among the sites’ tools is far from random. Different distances to resources may have led to differences in the degree to which they were exploited (Renfrew 1977). Consequently, more artifacts of nearby than of distant rocks would be present in a site collection. If the only factor influencing the choice of exploiting nearby or distant sources of identical materials had been distance, the closer resource was probably exploited more extensively. Exploitation of Nearby Lithic Resources In this section, the extent to which people depended on nearby resources will be examined and compared with exploitation patterns among sites of the same sector. In the western sites, BgDs4 and BgDslO, the levels of quartzite are 33% and 26% respectively (Table 2). Although quartzite usage was both relatively high and very similar in these sites, it does not constitute the bulk of all selected lithics. Usage of nearby rocks was neither exclusive nor consistent among the four eastern sites (Table 6) . The frequencies of black siltstone are relatively high in the majority of the sites, but the patterns of porphyritic tuff and black volcanic are very different. A high peak for each material is reached in two sites (BgDrll and BhDrl, respectively); their quantities in the remaining eastern sites are negligible. The uneven distribution may be indicative of some sites’ 67 closer proximities to quarry areas. Unfortunately, since specific quarries have not been pinpointed, this interpretation cannot be confirmed. Quartz, a material widely distributed throughout the Bay region, exists only in small amounts in most site collections (Table 2). TABLE 6. DISTRIBUTION OF EASTERN ROCKS IN EASTERN SITES. EASTERN SITES EASTERN ROCKS BgDrll BgDr7 BhDrl BgDr5 % (#) % (#) % (#) % (#) Porphyritic tuff 16.2 (29) 3.1 (1) 1.8 (1) 6.5 (7) Black siltstone 21.8 (39) 28.1 (9) 9.1 (5) 17.8 (19) Black volcanic 6.1 (U) 0.0 (0) 50.9 (28) 4.7 (5) TOTAL 44.1 (79) 31.2 (10) 61.8 (34) 29.0 (31) Few rocks other than quartz and quartzite suitable for making chipped tools were observed in the Perry conglomerate near BgDs4 and BgDslO. Both sites made use of each material to the same extent. On the other hand, the relative quantities of each of the four eastern rocks vary considerably among nearby sites. The eastern pattern may be indicative of 1) restricted natural distribution of one or more materials, 2) closer site proximity to one source than to another, or 3) more varieties to choose from than what was available in the west. Dependence on local resources in either sector was far from absolute—the proportions of nearby rocks in the sites’ collections range from 26% to 62%. Clearly, some means for obtaining raw materials found outside the immediate area of habitation existed. Exploitation of Distant PB Resources In the western sites, the proportions of eastern PB rocks, some of which may have originated in areas as far as 10 km away, is presented in Table 7. Selection for porphyritic tuff was minimal. 68 More black siltstone was brought into both western sites to supplement the local selection, but it was used to a greater extent in BgDslO. The frequency distribution of black siltstone in Tables 6 and 7 demonstrates a nearly equal rate of selection in all but two sites—BgDs4 and BhDrl. Increasing distance from the source of this material may explain its frequency fall-off in BgDs4, but does not explain its low frequency in the site nearest its source - BhDrl. TABLE 7. . DISTRIBUTION OF EASTERN ROCKS IN WESTERN SITES. WESTERN SITES EASTERN ROCKS BgDs4 BgDslO % (#) % (#) Porphyritic tuff 1.8 (2) 1.9 (4) Black siltstone 9.2 (10) 19.2 (41) Black volcanic 9.2 (10) 2.3 (5) TOTAL 20.2 (22) 23.4 (50) The eastern sites also made use of quartzite but its usage was negligible (Table 2). In summary, the occupants of the western sites made greater use of more distant PB rocks than did the people of the eastern sites, a reflection, perhaps of the presence of more varieties available in the east than in the west. Consequently, it does not seem likely that western site locations were chosen in order to minimize the distance to all major sources of PB lithic material. In only one of the eastern sites (BgDr5) was the quartzite level nearly equal to that of a nearby rock type. The degree to which PB rocks contributed to the total number of chipped stone tools is summarized in Table 8. 69 I | | Z"X zx z— ZX Z-x Z-X Z-X \D on un CO CO r——1 m x—z r—l m CM x—z x—Z XZ ( 0 ) (51) (19) 5-1 X-z (31) (56) 5-4 z Q P C>0 bO PQ o m 00 on CM CM o 0 . 0 17.8 29.0 r—1 47.7 52.4 CM 1 1 1 z-X ZX Z-X z-x Z-X ZX z-x Z-x Z-x Z-x Z-x =£= PQ Pm Z-X Z“X z-x Z"X z-x Z-x Z-x zx Z-x z-X z-x z-x Z-X SITES nD o m r—l 35 r- r- f—M CM CM on CM xz X-z r—l xz r—l r— XZ ^Z O m m Hl O CM O r—l X-z x—z r—l r—l X—z XZ X-z x—Z T—1 co XZ cn co X-z THE CO p J p O- 1—1 CM On M bC on AMONG & g ai M Z*X Z"X Z-X ZX Z-x Z-X Z-x Z-x Z-x H ***-. nD CM CM 35 Xr= 00 O CM r- o o r* cn xz Ob CM X—z nD o ROCKS P M p bO Pd bO PQ o CM n- On O PQ m on 00 o o o • • • • PB 6< Pm 6^ O') r—l nD o o on M2 nD 00 o on co CM in Ph r—l 1—“1 OF O JS O HI E-i P PQ nO HI N CU CJ 8. t* W o E-i On 4-> G 6 G G 4-> H M >n TABLE H 70 With the exception of BhDrl, use of local lithics ranges between 50% and 60%. BhDrl has 73% PB rocks. Utilization of Imported Material The presence of large quantities of nonPB rocks in most sites suggests that reliance on PB resources was nearly equalled by that for foreign materials in most sites. Within each site, there is considerable variation in the relative frequencies among foreign rock types (Table 9). Cryptocrystalline quartz and green mudstone are the most common types. Differences in desirability, tradition, resource distance, or other factors affecting acquisition could account for intrasite variation in the amounts of different foreign materials. Among sites, the proportion of each foreign material varies little. In summary, lithic raw materials were obtained from local and distant Passamaquoddy Bay sources, although the extent to which the latter were exploited may have been dependent on what was available in the immediate vicinity of each settlement. Utilization of PB rocks was usually matched by that of foreign materials. There is no evidence to suggest that settlement location was dependent on the availability of lithic raw material. People may have chosen an area to live for reasons other than the presence of suitable lithic materials, then utilized rocks they found nearby. In most site assemblages, the combined amounts of distant PB and imported materials is much greater than the amount of locally available rocks. 71 TOOL CLASS AND RAW MATERIAL SELECTION A feature common to many stone tool assemblages is that different kinds of raw materials were chosen for making different kinds of tools. For example, most tools from the Fernaid Point site on Mt. Desert Island, Maine, were made of local rock. Unifaces were \ manufactured of nonfelsitic materials, whereas felsite was used more often in biface production (D. Sanger 1980:29-30). Selection of lithic material was also different for unifaces and bifaces recovered from the Goddard site in Penobscot Bay. Here, local rock was used for making bifaces, but scrapers were made of imported chert (Bourque and Cox 1981:16). If the proportions of unifaces and bifaces are not the same in a site, then material selection patterns based on an entire assemblage would be skewed in favor of the larger tool class. As seen in Table 10, the number of bifaces of typed material is much greater than that of the unifaces in all but one PB site (BgDr7). The proportions of all unifaces and bifaces removed from each site vary by less than 3% from the figures in Table 10. TABLE 10. UNIFACES AND BIFACES OF IDENTIFIED MATERIALS: SITE TOTALS. UNIFACES BIFACES TOTAL SITES # (%) # (%) # BgDs4 36 (33.0) 73 (67.0) 109 BgDslO 49 (23.0) 164 (77.0) 213 BgDrll 57 (31.8) 122 (68.2) 179 BgDr7 23 (72.0) 9 (28.0) 32 BgDr5 46 (43.0) 61 (57.0) 107 BhDrl 16 (29.1) 39 (71.0) 55 695 72 Exploitation of PB Resources for the Manufacture of Unifaces and Bifaces Distance to a PB resource often produced parallel patterns of material selection among the sites’ unifaces and bifaces (Tables 11 and 12). For instance, quartzite levels are highest among unifaces and bifaces in the sites closest to its source—BgDslO and BgDs4. A similar pattern is evident in the assemblage from BgDrll, where more porphyritic tuff was used in both tool groups than in any other site. Black volcanic levels are highest in the bifaces and unifaces from BhDrl, which may have been located near this material’s source. The frequency drop-off for black siltstone bifaces in the westernmost site (BgDs4) is not paralleled by a similar drop-off for black siltstone unifaces at the same site. There is correlation between some materials and tool type. Black siltstone, black volcanic, and porphyritic tuff/rhyolite were chosen more often for making bifaces. The friable nature of quartz taken from the Perry conglomerate, as well as its small size in the glacial drift, may have restricted its usage in the manufacture of anything larger than the size of a uniface. Quartz unifaces are rarely well-made, and are frequently formed on jagged chunks of material. Selection of Imported Materials for Unifaces and Bifaces The relative percentages of imported rock types within and between tool groups were quite different (Tables 13 and 14). For unifaces, the most abundant import by far was cryptocrystalline quartz; this material appears in insignificant amounts in most sites’ biface inventories. Like quartz, its restrictive size may have been partially responsible for its higher frequencies in unifaces. 73 | 1 | Z~X Z“X Z~X Z~s Z-*x Z**X Z-S Z-~X z*^ z~x =»fc 00 ax i—4 in X^Z x-Z X^Z x_z X^Z X_z m X_Z (1) r—4 x^Z (23) p P (33) p P bO 00 PQ r* ox 04 00 04 1.6 r—4 ox 1 1 1 Z~X Z-S z**s Z"*X z**x Z"X z*x z-x Z-S z*x z^x Z^s =«= 04 on t—4 o in 1—1 =>T= o t—4 o m on Ox 1-4 x^z ^z x^z z x_z r-4 r-4 X_Z ^-Z ^-z ^-Z x«Z 04 04 P x_z p x^Z P p 43 X PQ m 00 on o on ax pa o X© o 00 o B< B'S 04 oo x© o i—4 00 o 04 o 04 Ox Z~X z*x z“x Z-N Z~x z—x Z^ Z—s z^ Z*s Z^K Z-x Z-X * =r$= x© on o O'- o X© =$£: o o i—4 04 o on r* x^Z s-z x-Z x^z x^Z x^z r—4 ^z x—Z ^Z X^z x^Z P x_Z P P • P bC m 00 CO PQ r—4 o o <• o m H co pa o o i-4 04 o on SITE B'S H B< • • • • • • 1—4 M x© on o o o ox co [d o o 04 o on 04 i—4 on X© r—4 04 on H 33 H O EACH H C H Z^ z-*s Z~X z**x Z"X Z*s Z"X M Z“*\ z*\ Z-S ✓—X Z*x Z**s z—X co :=*T= x© 04 x© 04 o CO =$>= 04 04 m on Ox i—4 r—4 x_z V—Z x_z x_Z x^Z x^z 04 s 1—1 V^-Z Z '^-Z 04 on X-Z O' i—1 o r—4 X—Z X^z x^Z FROM p OS p p P P bU in m o m m o bO X© X© m o r—4 PQ B'S co pa B^ o on t'- o on m M i—4 i—4 o n> 00 1-4 1—1 on O 04 04 m < P H Z~X Z"X UNIFACES z-“x z~x Z-S pa Z-S Z*s Z-S z-*x Z^x ^X r—4 r—4 i—4 =*r= pa B'S pq B-S 00 04 o 04 o on co O i—4 04 r—4 on 00 i—4 X© J 04 04 B< BS t—1 on o i—4 in i—4 P o 04 04 00 i—4 z o 1-4 H 4-1 P P IP a) pa IP 0) 3 a CJ P 3 3 CJ P o •rl as P O •rl p 3 H P 3 CJ CD 3 co CJ CD 3 •H P CJ 1-4 •H P CJ r— r—1 r—1 DISTRIBUTION CO o •rl p CD < C > o N N >x H • o N N Pn H ©3 P P X! ^5 O 04 P 4J P X X X O 11. P P d o CJ H r—4 P P Cl CJ CJ H nJ 05 P nJ ctf nJ nJ P 3 3 3 3 o r—1 1—1 M 3 3 o 1-1 rd O' O' p pa pa P er er p pa pa pa C TABLE H 74 | 1 1 1 | z-x z— Z"X Z-'x Z~x Z-X Z~x Z~X Z-x /—X z-s =t|=: r—1 CM =*$= O> r—4 CO r—4 o 00 (0) (0) LC CM (0) xz (0) in '^Z r—4 x^Z xz xz XZ CM P (23) P x-z s— P P 00 00 PQ r^- CO PQ \O r—1 a> KO O 00 r—4 r—4 0.0 O.o m 0.0 MT 0.0 so i—4 CD m I 1 1 1 1 z—s z~x Z~X Z*X z-x z*s Z"N Z-X Z-*\ z^ Z*x Z-x Z-x m o o o o o in co • z-x Z*X Z"X Z*X z*x z-x w /-X Z—X Z—X sO o T—-4 o o o r- E-i co r—4 co CM o CD sO XZ xz xz xz xz XZ .z xz ^Z x_z XZ' XZ x^Z XZ XZ SITE I"- H r-' P cn P P P OC P 0£ cn PQ r—4 o co o o EACH B-S c B'S r—1 M sO o o o o cd M M co o CM o CD sO CM co co t—4 CM sO H SS E-i O FROM 1-4 Dd l—1 Z~x z-x /*—X Z““X Z~X Fn Z*\ Z-X z^ Z*x Z-x Z-s Z-x cn CO i—1 o CM r—4 o CO co CM CM co r—4 r-H XZ CO xz xz xz x_z XZ co co r—4 •m_Z x^Z CM x-z r—4 XZ XZ m r—4 XZ XZ W r—4 X—✓ X-Z XZ p CD p p m CM CD r—4 xz xz co ^■4 X^ x-z x^ xz CD p CD a < P sO o o r--1 o Zs Z"X z-x Z"X Z*X Z~X S3 Z*\ z~x Z-x Z-^ Z-x Z-x Z-X Z-X =TT= CM r—4 cd r—4 o o OF CO co o CM sj S3 o p H P PQ nJ H nJ N CU U CU Dd N CU O CU 4-1 C •rH nj CD E-i 4-» PS •H nJ CD P o CU p CU cO C/2 P o cu a CU CO cO +-> p CO CD 4-1 H Cti 4-> a CO CD 4-> CZ) P CD o CJ e nJ H • o • e nJ > e nJ H Dd o P CU CD <• Dd o 2 cu O 13. 4-* P e p p 4-> H r—4 4-> a 6 a d 4-1 E-i 4J O- CU CU S 4-1 P cu cu s Po (U nJ CU i O w z^» cu nJ CU I O p (0 p 0) P p p cu p CU P §p P o o ai O p co o e Dd O P CO TABLE E-4 75 Stemmed and nonstemmed bifaces of cryptocrystalline quartz recovered from the PB sites are relatively small. Green mudstone seems to have been imported expressly for the production of bifaces, and is the dominant import in this tool class. Large quantities of its detritus (some of which are little-worked cobbles) have been observed in the collections and demonstrate that importation was not limited to finished or nearly-finished products, or that biface curation took place at the sites. That few flakes of cryptocrystalline quartz have been observed may indicate the importation of finished goods, the generation of fewer or small flakes from a smaller material mass, or the loss of small manufacturing or curation flakes during screening. Compared to the green mudstone and cryptocrystalline quartz patterns, the differences in import utilization for red mudstones and green volcanics are less marked. More appear in the sites’ bifaces than in their unifaces. Due to its extremely low frequencies, interpretation of ferro-manganese metasedimentary rock selection is more tenuous. It may have been used less often in uniface than in biface production. White-spotted metasedimentary rock appears only in bifaces from two sites. In conclusion, there seems to have been little variation in the rate of material importation between the two classes of tools from each of the sites (Table 15). Two rock types—cryptocrystalline quartz and green mudstone—were brought into the sites at a much greater rate than were others. Like the PB rocks, some imports were chosen more often for making one kind of tool. For no imported 76 material, however, is a peak reached in both bifaces and unifaces from the same site. Pattern Integrity Within the Biface Group The bifaces can be divided in four formal groups: stemmed, thin and nonstemmed, stemless fragments, and thick stemless, possibly unfinished bifaces. The relationship of each group to another is one that is not easy to define. The large, little-worked bifaces may be either or all of the following: cores, unfinished stemmed bifaces, unfinished thin nonstemmed bifaces, or completed tools in their own right. Nonstemmed bifaces (NSBs) that are thin are more likely finished or nearly-finished products, but they may also represent incomplete stemmed bifaces (SBs). Stemmed bifaces are finished or nearly-finished tools. A biface medial or tip fragment once could have been part of a SB or a NSB. Had these tools been used for different purposes, materials with different qualities may have been selected for their manufacture. Overlapping or mutually-exclusive patterns among the subgroups would thus correspond to either unity or division within the biface category. Comparison of stemmed biface and thin nonstemmed biface material selection. These tools are distinguished by differing treatments of the basal area of the tool: stemmed bifaces have hafting elements whereas nonstemmed bifaces do not (Plates 15 and 17). The imported 77 Plate 15. Thin nonstemmed bifaces from the Carson site 78 green volcanic seems to have been used more often in making stemmed bifaces; red mudstone and the iron-manganese metasedimentary rock appear more often in the nonstemmed bifaces (Table 16). For the most part, the patterns of material selection are so similar that differences at least in form are not accompanied by significant variation in the sets of chosen materials. Biface fragments and material selection. If little variation between the material selection patterns for SBs and NSBs has been observed, then that for the biface fragments should be comparable. The general pattern that emerges is one that is not very different from the complete bifaces (Table 17). Thick NSBs and material selection. As in most biface subgroups, black siltstone and green mudstone share the highest counts among the sites (Table 18). Quartzite usage is high in BgDsA, but unexpectedly low in BgDslO. The relative frequencies of the remaining materials do not differ greatly from their relative frequencies in the other bifaces, and suggests that the large bifaces could be uncompleted stemmed or thin nonstemmed bifaces (Plate 16). The alternative, that they represent the toolmakers’ original intent, cannot be ruled out. In conclusion, variation among the biface groups is minimal; large discrepancies may be attributed to extremely small sample sizes. Tool Class and Additional Raw Material Characteristics Raw material physical characteristics besides size and form seem to have influenced the choice of rocks for making unifaces and bifaces. For instance, of the PB rocks, those chosen consistently for uniface manufacture were quartz, quartzite, and black siltstone—all tough, hard materials with a high silica content. In crypto- 79 2 r> m o o o xf XI 'ZJ X in in o CM r i Kt 2 1 1 CT* CM co >n xJ o o a itJ z-x z-x Z~x z-x z-x z-x ZX z-x zx Z-X Z-x z— z-x «- — — ni o — .o o r< o o "r XJ <—> - - - - CT* ? , P* b~* i » xJ J* — xT P* z-x z-x zx zx. z-x Z-x ^x ,-x * o — o r j •jo — .n CM o o o jO — tn CT* ao C* -C X) W 1 1 1 , 1 o xT o O SITE. o o o rxj <—i — a o o o cm cc tn z-x z-x z-x ,-x Z-X z-x z-x Z-x z-x z-x Z-X M. o o o —• o — o o o o xl XJ cn z o o o o 1 1 1 1 1 1 LEVELS n o o CM in o “• aa zx z-x Z-x z-x z-x z-x. z-x Z-x z-x Z-x z-x o o o o o o o — o o tn SQ tn O o o MATERIAL Ixi W 1 1 1 1 1 1 1 i 1 o o O in in o z-x z-x z-x z-x z-x z-x Z-x Z-x z-x (NSB) o o CT* XJ tn X CM O' m CT* c-i o w 1 1 O CM £ o CM in J-X c> n o »n o o CT* CM 00 tn CM ex aO CT» CM CM - Kt 1 1 1 zx ^x z-x z-x ^X Z-X z-x Z-x ^x NONSTEMMED n o CM ao c—» xO X cn cn z xO x> xO CT* ex O ao 'XJ - THIN Kt m m xO o in o CM o AND 0£ Zx z-x z-x Z-X Z-X Z-x Z-X Z-X Z-X CO o m in in CM o CM CM XJ '— tn (SB) in co t xO - in in - - - - Kt CM z-x. z-x. zx Z-^ z-x *- o o m o — — o O o BIFACE X tn z O O O o o o o Kt 1 t 1 1 1 xT o o »n in in in o OJ CM CM cn o O STEMMED co o CM o CM O CM o o XJ OF cn r-» CM CM 7 - - - K< 1 1 1 1 1 33 ct* O' ao CO o o V “O c u u V O C x> A M C u G U -o > 8 -o o O P o c H c c 16. at u> U a TABLE flj N3I3H03 80 ; 1 i z—x (I) z-x z“x Z-x co 1—1 i—4 m o o M) £ PQ 00 co CO CO O o CO ’ 9 0.0 0.0 00 r—4i 31.3 CO o r-4 1 z~x Z~X Z~X z^x z~x z~x Z^x Z~X Z*x Z~X z-x ^X o o o o o i—4 o f—4 o o o CM i—4 X_Z x_z XZ X__ ' 1—1 x—' XZ x—z ^_z x^z x-Z i—4 X_Z x«z P PJ PQ o o o o co co o co o o CD Oc B< o o o o co 00 o 00 o o O Ox 00 OX =£= o o o 1—1 o o o o 1—4 o o CM r~ X_Z x__z ^Z x^Z x^ xZ XZ XZ ^z X--' x^z 04 P MJ CO PQ o o o o o o o o o o CD o B-S w o o o o o o o o o o CD o m m o H 1—1 H z^\ z-x Z~x /—X Z~X z~x z~x Z~x Z^X Z~X Z^x Z^x Zx, CO i—4 o tn OX r-4 CM Z~X Z*X Z—X z-x o CM 1—1 CM r—4 r—4 <■ CO DISTRIBUTION PQ CM CM CM OX kD 00 o m in o o o CM CM CM CD i—4 Z“X z~x Z~x Z“X z“x Z~X Z~x Z—X z~x Z^x z~x z-x Z-X =$£= cd uo o CM MATERIAL cn P MJ PQ o r- o t—l CM i—4 co m r-4 CD CD o B'S o CM o ox 00 o> p~- <■ Ox O O o Csl i—4 CM o r-4 FRAGMENTS: BIFACE co 4-1 4-> •H o U) o • 6 nJ > e nJ H o N N • o z> 17. nJ 04 ns nJ >> CD u 9d N0I3B0I TABLE 81 L J ? I 1 1 1 1 r~ r Z-X Z-x Z-x (I) r—4 r—1 Cxi o x—z (0) in xz (0) (2) XZ (0) (0) xz (0) X-z (7) O 00 € PQ CO CO o ‘ VI 0.0 O.o 0.0 28.6 r—4 r—4 CM 100.1 1 ) Z—X z—, Z-x o o o r—l CM o O o o o o co r—1 XZ X—Z XZ XZ xz X-z XZ XZ X—z xz X—z xz xz 5-4 p 3= PQ o o o CO o o o o o o B'S • o o o co io o o o o o o cn co z-x Z"X Z-x Z-x Z~x z-x Z—X Z-x z-x zx Z-x z-x o o r—4 o o o O o o o o r—"4 r- X—z XZ x^z XZ X—z X-Z xZ XS' XZ xz xz X—z XZ 54 P t>0 cn PQ o o o o o o o o o o o o B'S • W o o o o o o o o o o o o o c H r—4 r—4 r—l z-x z-x Z----s zx ✓—x uo r—4 o CM r-~ o o o r—4 o o o r—“4 xz XZ xz XZ X----' x—z X-z X^z xz xz xz xz CM i—■1 XZ 5-1 P 00 o o o o o o o o o o o o PQ B'S m o o in o o m o m o o o r—4 co Z^X z-x Z-x z-x z-X Z-x Z-x Z-x Z-x Z-x Z-x Z-X z-x JX. o in r—4 00 r—4 co !—4 CM o o o r—1 O XZ X-z XZ ■—z xz x-z r—4 X-z xz x-Z co r—1 XZ - ----' <0 P 00 o r—4 CM 00 CM in m o o o o PQ B'S • o i£> CO z-x zx z-x =£= o cm O o O co xt o r—4 o o o a) Ti N CD BIFACES: a o u CD •H CD o •H 4-J S nJ 4J s 5-1 O 18. nJ cO 5-i cO CO X CD nJ CD 1 o O 1—1 T----1 5-i 54 CD 54 CD p Q> Q> O' O' PU PQ PQ O O PQ O P cn Sd N0I3H03 TABLE 82 Plate 16. Thick nonstemmed bifaces. 83 crystalline quartz, several characteristics found in all, but not in any one PB rock, are combined. Unlike quartz which, because of internal fractures, breaks into smaller pieces, cryptocrystalline quartz is often less friable and is probably easier to control during reduction. It is finer-grained and may be harder than quartzite or black siltstone. It is also more colorful than any rock outcropping in the Bay that was used for making unifaces. The greatest usage of cryptocrystalline quartz occurred in BgDrll and BgDr5 (Table 13). These people were not as dependent on nearby resources as were the people in other sites. Materials suitable for uniface manufacture may have been absent from the local resource selection for BgDr5 and BgDrll. The indigenous raw material that was used most often in biface work is black siltstone; quartzite, porphyritic tuff, and black volcanic were also frequently chosen in sites near their sources. Although patterns of distant PB rock selection varied considerably between sites of western and eastern sectors, their dependence on foreign materials did not. The import with the highest site percentages is green mudstone. It, along with red mudstone, has one characteristic not found in PB rocks that were used in making bifaces: a tendency to split along remnant bedding planes. They are also finer-grained than any PB material frequently selected for making bifaces, and are more colorful. Compared to those of the unifaces, more of the biface materials are less silica-rich and are larger-grained. Many lack a homogeneous structure, having either phenocrysts or cleavage planes. In some, conchoidal fracture is less pronounced than in most uniface materials. 84 Materials with these characteristics may have been considered less suitable for the manufacture of unifaces, or for their function. The materials once considered to have superior flaking or functional qualities need not have been the ones chosen most often for making tools. Custom may have dictated the procurement of rock that was inferior to others (Gould 1980:149-155); lower quality materials may have been obtained with greater ease. Finally, although rocks are often ranked according to their physical qualities (Fitzhugh 1972:106, Gould 1980:217-219), we cannot presume to know which materials were at one time considered "superior" to others. TEMPORAL PATTERNS OF RAW MATERIAL SELECTION Intrasite and intersite temporal variation in Aceramic and Ceramic material selection can be examined in the assemblages from the Minister’s Island (BgDslO) and Teacher’s Cove (BgDrll) sites. Clear-cut temporal associations of all artifacts based on stratigraphic placement in these sites is not always possible; separation of Aceramic and Ceramic unifaces and stemmed bifaces is accomplished more often through stylistic attributes. Most of the PB stemmed bifaces can be linked stylistically to those in regional sites that have been radiocarbon dated or have better stratigraphic control. A brief description of the diagnostic point types from each time period follows; for a more detailed description of points from the Teacher’s Cove site see Davis (1978:19-21). The Temporal Framework In two Passamaquoddy Bay sites, BgDslO and BgDrll, several stemmed bifaces generally resemble Snook Kill or Atlantic points of 85 New York and Massachusetts: large, broad, and contracting- or parallel-stemmed, with convex or straight bases, often with pronounced angular shoulders (Plate 17A,B), (Dincauze 1972; Ritchie 1961, 1969a). This style has a wide distribution throughout the Northeast, occurring in both interior and coastal sites such as Hirundo (D. Sanger, et al. 1977), Ellsworth Falls (Byers 1959) and Turner Farm (Bourque 1971) in Maine, and Bear River in Nova Scotia (Connolly 1977). At the Young site, located near Bangor, Maine, associated radiocarbon dates range from 3715 ± 60 to 3105 ± 50 B.P. (Borstel 1982:64). These dates conform to Ritchie’s date of 3470 ± 100 B.P. from the Snook Kill site in New York which has similar artifacts (Ritchie 1969b:136). As part of the Susquehanna/Broadspear tradition, such stemmed bifaces predate the utilization of pottery in the Northeast. Another large Aceramic projectile point that may not have had as wide a geographic distribution is one that was broadly notched with a convex expanded base often narrower than the width at the shoulders (Plate 17C). This style is present in the stemmed bifaces from BgDslO and BgDrll. Similar stemmed bifaces have been found at the Young site in the same level as Vinette I-like pottery (Borstel 1982:20;22, Plate 4:T), and at the Bear River site (Connolly 1977, Plate 2). They have also been recovered from BgDr8, located near the Orr’s Point site, and the Goddard site on Penobscot Bay, which lack the older ceramics. The Goddard point was excavated from a charcoal-bearing pit dated at 2840 ± 105 B.P. (Bourque and Cox 1981:11-12, Plate 1:9). 86 size) Ceramic straight D, actual Middle Nonstemmed Ceramic 7/10 H, shown Middle F, broadly-notched; corner-notched; (Specimens shallow-notched; Aceramic C, Ceramic Rossville-like; Ceramic Ceramic) Late E, I, Late Middle J, G, Atlantic-like; B, stemmed; and biface: wide-notched; Meadowood-like; 87 One Meadowood-like point, possibly of material foreign to Passamaquoddy Bay, has been identified in the assemblage from Teacher’s Cove (Plate 17D). It has the characteristic Meadowood attributes: finely-made, small side notches, and a wide neck above a rectangular stem with a ground, slightly concave base. In New York, these stemmed bifaces have a time range of about 3000 to 2500 B.P., and are often associated with Vinette I pottery (Ritchie 1969b:180), which is not present at Teacher’s Cove. At the Young site, a few Meadowood-like points have been found in the same context as pottery resembling Vinette I (Borstel 1982:22, Plate 4:A-C; 24). The Goddard site (Bourque 1971:Figures 4, 105), the St. Croix Island site (D. Sanger 1973), and an interior Nova Scotia site (D’Entremont and Moore 1977:34, Plate VI) have yielded similar stemmed bifaces. Several examples of Rossville-like stemmed bifaces appear in the assemblage from BgDrll (Plate 17E). The type specimens from New York are: ...roughly rhomboidal or lozenge-shaped (with) weak, oblique shoulders which merge with a contracting stem terminating in a blunt point (and) occur in the lower levels of certain coastal New York shell heaps, apparently without pottery associations [Ritchie 1961:46]. Rossville points also occur with pottery. On Martha’s Vineyard, they are in context with Vinette I, and have been dated at 2470 ± 120 to 2050 ± 80 B.P. (Ritchie 1969a:205, 244). Farther north, they have been identified in the Young site (Borstel 1982:22, Plate 4:G; 25-26), Moose Island site (Kingsbury and Hadlock 1951:Plate 1), and at the Sand Point site (BgDs6) on the eastern shore of the St. Croix River (Lavoie 1972). 88 In Passamaquoddy Bay, the first indications of shellfish exploitation are dated at about 2000 B.P. as are the first known ceramics. Much thinner than Vinette I, this pottery is tempered with grit and decorated with rockered and stamped dentate designs. It is found in the three oldest sites under study: BgDrl1, BgDslO, and BgDs4, as well as in the Sand Point site (BgDs6), with stemmed bifaces quite different from points of the Late Archaic/Early Ceramic (Aceramic) cultural period. Middle Ceramic (2000-1000 B.P.) points are much smaller in size and are usually more finely made. Three forms have been distinguished. One is a straight-based biface with a parallel stem much narrower and longer relative to the blade length than those of the Aceramic stemmed bifaces (Plate 17F). It occurs with early dentate ceramics at the Turner Farm site (D. Sanger, personal communication 1982). Another Middle Ceramic point has broad, shallow, nearly indistinct side notches and weak shoulders (Plate 17G). Lastly, there are points with wide, pronounced side notches, angled shoulders, and straight or slightly convex bases which are nearly as wide as the shoulder width (Plate 17H). By 1000 B.P., the number of dentate-decorated ceramics had declined as a different kind of pottery became more popular. Thick and shell-tempered, it is impressed with a cordwrapped stick design. During the Late Ceramic period (1000-500 B.P.), the associated stemmed bifaces became thinner with small corner and side notches, shallow serrations, well-defined shoulders that are often very narrow, and thinned straight or concave bases (Plate 171) (D. Sanger, personal communication 1975). There are some indications that the neck narrowed with time. Similar pottery and stemmed bifaces are found in 89 all the PB sites, as well as in numerous sites throughout the Maine- Maritimes region with comparable dates. Compared to the bifaces, the Passamaquoddy Bay unifaces cannot be fixed as securely in a temporal framework. Aceramic period unifaces are generally much larger than their Ceramic counterparts. For instance, Kingsbury and Hadlock distinguish their younger small thumbnail type scrapers from large unifaces found at the base of a shell midden on Moose Island in nearby Cobscook Bay, Maine. They equate the large unifaces with those found in the lower levels of other Maine shell middens, including the Taft’s Point site in Frenchman’s Bay (Hadlock 1939, Kingsbury and Hadlock 1951:25). Among unifaces excavated from the Teacher’s Cove site, Davis discovered a trimodal weight distribution that apparently corresponds to their temporal placement (Davis 1978:28, 29). Many of the largest exceed 10 g, and are associated with Aceramic stemmed bifaces. Of the six PB sites, the two with the greatest proportion of unifaces exceeding 10 g in weight also contain well-represented Aceramic components (BgDrll and BgDslO) (Table 19). TABLE 19. PROPORTIONS OF COMPLETE UNIFACES <10g OR >10g IN EACH SITE. SITES BgDsA WEIGHT BgDslO BgDr 11 BgDr7 BhDrl BgDr5 Z (#) Z (#) z (O Z (#) Z (#) Z (#) £125 87.8 (36) 76.9 (40) 69.8 (44) 96.3 (26) 94.1 (16) 95.5 (42) >10g 12.2 (5) 23.1 (12) 30.2 (19) 3.7 (1) 5.9 (1) 4.5 (2) TOTAL 100.0 (41) 100.0 (52) 100.0 (») 100.0 (27) 100.0 -U-U- 100.0 (44) 90 The presence of large unifaces in the younger sites may be attributable to the tail end of a normal distribution of smaller Ceramic unifaces or, possibly, to the presence of a remnant Aceramic assemblage (as in BgDs4 and BgDr5). The weight distributions of unifaces from each site are presented in Figure 8. In the two multicomponent sites, the distribution extends far beyond the 10.1 g mark. With time, the distribution shifts toward smaller-sized artifacts as the range becomes more restricted. Since the upper end of the curves’ distributions in at least three of the younger sites (BhDrl, BgDr7, and BgDr5) terminates at or near the 10.1 g mark, this point is used for separating the unifaces of the oldest sites. It is recognized that the utilization of weight as a distinguishing attribute is an arbitrary procedure that is not without problems, but better criteria for sorting Aceramic and Ceramic unifaces in sites with mixed stratigraphy have yet to be formulated. Few of the nonstemmed bifaces could be assigned to either time period. In the Maine-Maritimes region, David Sanger has observed more large bifaces in Aceramic than in Ceramic sites (D. Sanger,* personal communication 1976). Accordingly, several attributes of NSBs from Passamaquoddy Bay’s single and multicomponent sites were examined for temporal exclusiveness. No temporal patterning was seen in plots of width, thickness, and length measurements, or in artifact and basal shape. Because of poor stratigraphic control in BgDslO and BgDrl1, it is not possible to distinguish Aceramic from Ceramic NSBs by their vertical placements. 91 ------7 Aceramic/Ceramic BgDs-10 n = 52 Figure 8. Weight distributions of complete unifaces from each site 92 Two NSBs from the Minister’s Island inhumation are attributable to the Aceramic period through their association with artifacts having established temporal identity, through stratigraphic placement, and by associated radiocarbon dates. One of these, a large oval biface, is quite unlike any other found in the sites’ collections, but its role as a burial good may preclude comparison with the utilitarian items. The other NSB is small, thick, asymmetric, and lanceolate. Both artifacts are of materials probably not native to Passamaquoddy Bay (one is of white-spotted metasedimentary rock, the other of quartzite), as are two of the burial’s three stemmed bifaces (red mudstone, white spotted metasedimentary rock, and black volcanic). One artifact recovered from BgDr5 is a small, concave-based, nearly-equilateral triangular biface not unlike the Late Woodland Madison point identified by Ritchie (1961:31) (Plate 17J) . That more have not been found in the Bay area may be a function of their geographical distribution in the Maine-Maritimes region. Sanger, for instance, has observed that while notched points have been found in Late Ceramic sites of eastern Maine, more of the triangular form are found in the central and western part of the state (D. Sanger 1979:113-114). Traditions of Uniface and Biface Material Selection Stratigraphic variation in frequencies of raw materials has been documented in regional archaeological studies. Sanger observed a material shift in detritus—from felsite (here referred to as green volcanic) in the Aceramic levels to aphanitic materials in the Ceramic strata—at the Frazer Point site on Schoodic Peninsula (D. Sanger 1981a: 35). A late preference for cryptocrystalline quartz and other 93 aphanites, was also observed in unifaces and flakes from the Young site (Borstel 1982: 40, 45). In Table 20, material percentages in unifaces from both components of BgDslO and BgDrll are compared. Unifaces associated with the sites’ Ceramic components are far numerous than those with the Aceramic components. Nonetheless, as best as can be judged from the small samples, there are few temporal differences in PB material usage within each site. Foreign material selection varies little through time as well. Of particular interest is the high frequency of cryptocrystalline quartz in both time periods. Contrary to observations on many Maine sites, there is no support for an increase in its usage through time in these PB sites. Compared to the unifaces, there seems to be more intercomponent material differences in the sites’ stemmed bifaces (Table 21). More black siltstone may have been used during the Ceramic period, and white-spotted metasedimentary rock is present only in the Aceramic components. Although a time span of over 2000 years is represented in the radiocarbon dates from the Minister’s Island site (Teacher’s Cove was also settled during an equally lengthy period of time) (Figure 6), uninterrupted occupation of either site probably did not take place (D. Sanger 1981b:38). Any temporal pattern of lithic material selection based on the findings from BgDslO and BgDrll would be only partially representative of the Bay’s occupational prehistory. However, when dates from the four remaining Passamaquoddy Bay sites under study are added to the temporal framework, nearly continuous 94 TABLE 20. TEMPORAL COMPARISON OF COMPLETE UNIFACE MATERIALS IN BgDslO AND BgDrll. BgDslO BgDrll ACERAMIC CERAMIC ACERAMIC CERAMIC ROCKS % (#) % (#) 7o (#) 7o (#) Quartz 0.0 (0) 8.1 (3) 15.4 (2) 9.5 (4) Quartzite 50.0 (5) 40.5 (15) 7.7 (1) 2.4 (1) PQ Porph. tuff 0.0 (0) 0.0 (0) 15.4 (2) 4.8 (2) PU Black siltstone 10.0 (1) 13.5 (5) 7.7 (1) 9.5 (4) Black volcanic 0.0 (0) 0.0 (0) 0.0 (0) 4.8 (2) Crypto, quartz 40.0 (4) 27.0 (10) 53.8 (7) 59.5 (25) JZ Green mudstone 0.0 (0) 0.0 (0) 0.0 (0) 2.4 (1) O H Red mudstone 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) M ai Green volcanic 0.0 (0) 8.1 (3) 0.0 (0) 4.8 (2) O Ph Fe-Mn metased 0.0 (0) 2.7 (1) 0.0 (0) 2.4 (1) Spotted metased 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) TOTAL 100.0 (10) 99.9 (37) 100.0 (13) 100.1 (42) TABLE 21. TEMPORAL COMPARISON OF STEMMED BIFACE MATERIALS IN BgDslO AND BgDrll. BgDslO BgDrll ACERAMIC CERAMIC ACERAMIC CERAMIC ROCKS 7 (#) 7 (#) 7o (#) 7o (#) Quartz 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Quartzite 0.0 (0) 17.6 (3) 6.7 (1) 10.0 (1) PQ Porph. tuff 0.0 (0) 5.9 (1) 20.0 (3) 20.0 (2) Pu Black siltstone 0.0 (0) 29.4 (5) 0.0 (0) 30.0 (3) Black volcanic 16.7 (1) 5.9 (1) 26.7 (4) 20.0 (2) Crypto, quartz 0.0 (0) 5.9 (1) 0.0 (0) 0.0 (0) IS Green mudstone 33.3 (2) 17.6 (3) 13.3 (2) 20.0 (2) U l-l Red mudstone 16.7 (1) 5.9 (1) 0.0 (0) 0.0 (0) M ad Green volcanic 0.0 (0) 11.8 (2) 26.7 (4) 0.0 (0) o fl-l Fe-Mn metased 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) Spotted metased 33.3 (2) 0.0 (0) 6.7 (1) 0.0 (0) TOTAL 100.0 (6)* 100.0 (17) 100.1 (15) 100.0 (10) ^includes two nonstemmed bifaces from burial 95 occupation of the Bay area from at earliest 2450 to at latest 250 years B.P. is represented. Whether dependence on a raw material (or a group of raw materials) remained constant or shifted through time should be detectable by scanning the raw material frequencies in tools from the oldest through the youngest sites. In Table 22, the percentage patterns in sites of the eastern and western sectors suggest a trend away from utilization of cryptocrystalline quartz in uniface manufacture, a trend opposite that seen in many Maine sites. Furthermore, quartz usage, at least in the eastern sites, does not seem to have declined with time, as had been observed at the Young site (Borstel 1982:40). The shift towards fine-grained materials noted by Sanger is not consistently represented in the Passamaquoddy Bay unifaces (Table 22). Two PB volcanics—both with phenocrysts—have different temporal selection patterns in the eastern sites. Porphyritic tuff is highest in BgDrll’s Aceramic component, but black volcanic reaches its peak in a Late Ceramic site—BhDrl. Additional inconsistencies are evident: quartzite selection does not seem to have diminished by the Late Ceramic, and the aphanitic black siltstone, varying little through time, peaks in only one Late Ceramic site. Unlike the unifaces, cryptocrystalline quartz is present only in Ceramic period bifaces (Table 23) . The absence of cryptocrystalline quartz in Aceramic stemmed bifaces cannot be explained by ignorance of its existence since this rock is present in Aceramic unifaces. A finer-grained, rare variety of green mudstone that is nearly cryptocrystalline in texture has been identified only in SBs of the Ceramic period. All of these rocks are fine-grained and homogeneous in structure. 96 1 1 1 | | r—l 6*89 p m 00 co sO O p b^ r—4 P p CM 00 sO II c PQ r—H r—< 31.3 31.3 31.3 100.2 w o w p E-< P r—l o IN u 1—1 p jr] 1—1 p O O W P T—1 jr] Cfl r—< in in i—4 o r—4 p 00 as p Q W p 6< co P E-i pd bO 00 o CO CM p 00 CM p Os II Hl <3 w PQ r—l sO CM CO as pl S P o FOREIGN cn W E-i Hl u AND OF W & o p O jr] r—l (fl o o O o o o o fv: p b^ • • • • • • 1—4 W b£ o o o o o o II O PQ m r—4 sO 0) X Pl pl o N (U O 0) b£ o •H P C •rd X (0 •rd p a P o CU a a) co cu M-l (fl cO cO p pl cO co p p M-l p o PQ a co o a co T—1 • e X > 0 X r—l H o N N • cO o 2 (U cO O Dd P P pi pi P p ps e pl a p P E-i P P P o u O p 22. cO CO P (0 cO E-i CU X) (U i o H O t—1 1—1 P (U p a) p 0 3 p O' O' P PQ PQ o O Dd o P co Rd NDI3B0J TABLE 97 s© 00 o cn Ci B'S CM CM os i—-4 n=39 BhDrl| m r- CM 100.1 J ERAMIC 1 | C r~- n) ZZ 'Z Z p< r—l cn r—l CM s© •H Q »-< r—l i—l s© LATE 00 co CM £ 99.9 1—4 33.3 r—l PQ cn CM s© n=9 32 | | <4-1 m M sO m 00 S© s© r—l s© OS s© 00 CD Q B'S CD 00 r—l r—1 Os CERAMIC <4-1 LATE MIDDLE/ •rl N | | | | 1 | | 42 W H r—< nJ CD cn r—l <1 P o o M n oo o o CD 10.0 20.0 20.0 30.0 80.0 PQ CM CM n=10 4J 100.0 CERAMIC BIFACES CD MIDDLE a 1 | ] o CD IN s 1 CD O O cn IB nJ B'S 4J M-4 s© cn cn S© s© s© •rl 20.0 26.7 i—i n=15 BgDrl m CM nJ 32 100.1 ACERAMIC C nJ nJ | 1 | | | a) CD MATERIALS g a) 0 Z o a> sO Os 17.6 i—4 r—l n=17 BgDslO CM m 42 o 100.0 CERAMIC LATE MIDDLE/ a FOREIGN nJ Pn | 1 1 | 0) nJ r—l a pi SITES 5 nJ o CD r* CM 00 CM r» CM 1—1 4-> CD CD o Q B'S CD CD (D s© CM m O U PB oo CM 00 00 00 u . o 99.9 32.9 m CM n=73 U nJ nJ PQ •rl <4-4 CERAMIC M-4 MIDDLE •rl •rl OF ESTERN | nJ 42 42 P cn r- cn cn C B'S • • • s© s© cn s© cn cn «—4 a) 3 i—4 I—1 f—4 cn n=6 r—1 4-> 4-J BgDslO cn 00 100.0 CD CD ACERAMIC nJ <4-1 <4-4 m | | | | | o O o DISTRIBUTION CD CD CD 4-> +-» 4-> CD CD CD •rl •rl •rl CD CD CD p! C a O o o quartz Foreign O PB metased o o tuff TEMPORAL siltstone CD CD volcanic metased CD volcanic mudstone i-4 r—l r-1 TOTAL PU PU mudstone Pu 0 0 0 ROCKS 23. Total Total nJ nJ nJ CD CD CD 1 1 1 Spotted Red Crypto, Green Green Fe-Mn Quartzite Black 1 Quartz Black Porph. 1 1 r—l CN m I | | | | | | Rd NOISBOd TABLE 98 By the Late Ceramic period, the medium-grained grey quartzite was in little or no use in the eastern sites, but any trend in western sites is not evident. While black volcanic usage seems to have declined with time in BgDs4 and BgDslO, it did not change in BgDrll, and reached a decided peak in the bifaces from BhDrl—a Late Ceramic site. There is no decline in porphyritic tuff in either sector aside, perhaps, for its absence in one of the youngest eastern sites. For green volcanic, no change in pattern is apparent; evidence for a decline in felsite usage that was noted elsewhere is not present in the Passamaquoddy Bay sites. There is also no evidence for a trend of changing dependence on the PB lithic resource base. The patterns of volcanic selection in the PB bifaces do not seem consistent with a regional trend for increased selection of finer-grained materials through time. Within the stemmed bifaces, at least, there is a temporal pattern for selection based on the texture of the volcanic material. The proportion of each volcanic variety in Aceramic and Ceramic stemmed bifaces is remarkably similar (Table 24). TABLE 24. DISTRIBUTION OF VOLCANIC VARIETIES IN ACERAMIC AND CERAMIC STEMMED BIFACES. ACERAMIC CERAMIC VOLCANICS % (#) % (#) Porph. tuff 27.3 (3) 31.6 (6) Black volcanic 36.4 (4) 31.6 (6) Green volcanic 36.4 (4) 36.8 (7) TOTAL 100.1 (ID 100.0 (19) The similarity does not extend to choices made for material texture. In the collections, each volcanic variety is present in two forms: materials having far more large phenocrysts than small, and materials 99 with small phenocrysts outnumbering or exclusive of those that are large. Frequently, the texture of the matrix varies accordingly. Volcanics with small phenocrysts and a finer-grained matrix were chosen more often for making Ceramic stemmed bifaces than were the same rocks with coarser texture (Table 25). Selection for finer- grained materials corresponds with a stylistic change from thick, large, stemmed varieties of the Aceramic period to smaller, thinner, Ceramic projectile points with delicate notches. Coarse materials may not have been as compatible with the greater precision required in making notched tools. TABLE 25. STEMMED BIFACES: VOLCANIC TEXTURE AND TEMPORAL ASSOCIATION. ACERAMIC CERAMIC TEXTURE % (#) % (#) Large phenocrysts 45.5 (5) 21.1 (4) Small phenocrysts 54.5 (6) 78.9 (15) TOTAL 100.0 (U) 100.0 (19) The volcanic texture-temporal correlation was discovered after the nonstemmed bifaces and unifaces had been returned to the National Museum of Canada. Given the SB/NSB parallels in material selection noted earlier, it is suspected that more Ceramic than Aceramic NSBs would be made of fine-textured volcanics. In addition, since unifaces of either period were made of fine-grained rock, it is doubtful that any change in choices of volcanic texture would be found. A hundred years ago, Matthew observed a change in material preference during the course of excavating the Bocabec site (BgDrl), located near the Orr’s Point site: 100 In this lower kitchen midden, which is connected with the hut bottom No. 1...there were many flakes of a dark brown petrosilex [probably porphyritic tuff] of a coarser grain than the black petrosilex [probably black siltstone] from which many of the weapon-points of the higher kitchen middens and hut bottoms were made. Quartzite rock was more largely used in the manufacture of weapons by the inhabitants of hut No. 1 than by the men who subsequently occupied the place [1884:17-18]. Bishop, who recently reevaluated the cultural materials from this site, determined the presence of two occupations—one of the Middle and another of the Late Ceramic periods—based primarily on pottery attributes (Bishop 1980:57-59). The shift from coarse to finer-grained PB rocks in bifaces that Matthew noted is identical to the pattern seen in other eastern PB sites of the same periods, and is consistent with the trend observed at least in coastal and inland Maine sites. The choice of fine-grained materials for making unifaces is, however, a practice as old as the earliest sites that have been excavated thus far in Passamaquoddy Bay. Stone tool manufacture ended following the arrival of the Europeans in the sixteenth century. New forms made of iron and other metals were acquired through trade with the newcomers. In the Passamaquoddy Bay region, the loss to the indigenous culture was not limited to stone tools, for with them were abandoned traditions of raw material procurement and manufacture that had existed at least 2500 years beforehand. COMPARISON OF THE SOURCES OF VARIATION IN THE RESOURCE EXPLOITATION PATTERNS The choices of resource exploitation were affected less by changing material preferences than by resource distance or tool 101 production or use requirements. Temporal variation in material selection is greater in the bifaces, with more fine-grained materials chosen during the Ceramic than during the Aceramic period. Among the PB lithics, the utilization of black volcanic rock and, to a lesser degree, quartzite, seems to have been affected most by distance to their sources. Each rock type was equally suited to making bifaces and unifaces in sites near their sources. Among the other sites, quartzite and black volcanic quantities are not as large. The indigenous material whose utilization was least affected by resource distance was quartz, which is naturally distributed throughout the settlement area. Variation between uniface and biface levels of quartz is very pronounced. The exploitation of black siltstone and porphyritic tuff was influenced by resource distance and tool class. Both rocks were used more often for making bifaces than unifaces and are most abundant in sites closer to their sources. Of the foreign materials, selection for tool type is most apparent in cryptocrystalline quartz (unifaces) and green mudstone (bifaces). Distance to sources of lithic materials foreign to Passamaquoddy Bay cannot be compared because their locations are not known. Hypothetically, if foreign materials originated in areas where similar rocks outcrop, then a case for the effects of distance on their exploitation can be made. Of the foreign rock types whose potential sources have been identified (the exception is green mudstone), the cryptocrystalline quartz and white-spotted metasedimentary rock areas are closest to Passamaquoddy Bay. Relative to red mudstone, green volcanic, ferro-manganese and white-spotted 102 metasedimentary rock, cryptocrystalline quartz levels were consistently high among the sites. Since green mudstone quantities parallel the amounts of cryptocrystalline quartz, it would not be surprising to find its source closer to Passamaquoddy Bay than were those of other foreign rocks. White-spotted metasedimentary rock utilization, however, remained low, confined to the Aceramic period. A selection change is detectable among stemmed bifaces, but not unifaces, of the Aceramic and Ceramic periods. 103 Chapter 4 CONCLUSIONS In the Passamaquoddy Bay region, the same rocks were often used for making different kinds of tools by settlers living in different areas and times. The amount that each resource was used, however, frequently varied among the settlements. This study demonstrates that there is both uniformity and variation in the way lithic resources were chosen and exploited. The effects that site-to-resource distance, tool form, and changing material preference had on raw material selection are summarized below. Factors that Affected Lithic Resource Selection and Exploitation Resource distance. Nearby rocks were used more often than more removed PB sources. The quality of the local selection may have influenced the procurement of distant PB rocks. People in the eastern sites, for instance, seem to have been less dependent on distant PB lithic resources than were the people of the western sites. The selection of PB rocks was nearly equalled by that of rocks foreign to Passamaquoddy Bay. The presence of large quantities of foreign material in a collection suggests that it could be obtained with relative ease, but a material’s value could have outweighed procurement difficulties. Tool class. Different rocks are associated with tools of different forms. Fine-grained uniformly-structured materials were preferred for making unifaces; for bifaces, coarser-grained and foliated rocks were also chosen. Tool size and form is sometimes affected by material size and form. For example, bifaces of 104 cryptocrystalline quartz are small, and quartz was rarely fashioned into tools other than unifaces in the PB collections. For performing a particular task, some of the materials are probably better suited than others. The tool classes used in this study are based on shape, but within each may be tools that had different functions. If tool function guided raw material selection, then patterns quite different from those observed in the analytical groups would be expected. However, if stemmed and nonstemmed bifaces have different functions, this difference is not paralleled in their material selection patterns. Resource distance and tool selection. Distance to PB resource areas had similar effects on uniface and biface selection patterns. Among both tool groups, the highest level of a localized rock is reached in the same nearby site. Increasing site distance from a lithic resource is often accompanied by a decline in material levels in both tool groups. Cryptocrystalline quartz and green mudstone were the most frequently chosen foreign rocks for unifaces and bifaces, respectively. Until the sources of rocks foreign to Passamaquoddy Bay are discovered, it will not be known if cryptocrystalline quartz and green mudstone were any closer to the settlements than were the sources of other foreign rocks. Time and tool class. Among unifaces of the Aceramic and Ceramic periods, the same rocks were chosen in the same amounts with few exceptions. Among stemmed bifaces of the Aceramic and Ceramic periods, there was a change in selection of raw materials. Both foreign and indigenous materials are included. Similar patterns in 105 Maine sites spanning the same time periods support a trend away from coarse materials as notched bifaces became popular. More research is needed to determine if the trend was duplicated in nonstemmed bifaces. Time and resource distance. There is no evidence supporting a significant change in dependence on local and distant PB resources, or on foreign materials through time. An increase in foreign rock levels associated with the Late Ceramic component of the Goddard site (Bourque and Cox 1981:15) is not evident in the late Passamaquoddy Bay sites (BgDr7 and BhDrl). In BhDrl, nearby rock constitutes the majority of all materials selected for its unifaces and bifaces. Several options for coastal faunal resource exploitation were practiced by the people of the Maine-Maritimes region (Sanger 1982). It is also evident that choices of lithic resource exploitation varied among coastal sites. Coastal Adaptation and Lithic Resource Exploitation The Late Archaic/Early Ceramic (Aceramic) occupations of Passamaquoddy Bay were already heavily dependent on adjacent resources and on PB rocks found farther away. To some archaeologists, evidence for greater foreign material utilization would be expected in collections from an area’s initial settlements, to be followed by evidence for increased exploitation of local lithic resources (Funk and Rippeteau 1977:37). This could suggest occupation predating the currently known sites. Site loss through erosion is likely. Despite changes in tool form, lithic material selection patterns observed in the Aceramic assemblages continued in the following Ceramic period in Passamaquoddy Bay. The same varieties of rock were 106 used, and the proportions of each may have fluctuated only in accordance with a regional shift in material preference accompanying the stylistic change. It would not be surprising if continuities in subsistence strategies are discovered. Minor changes in projectile point styles, pottery manufacture and decoration techniques, and subsistence (McCormick 1980:112) during the Ceramic Period have been documented. Patterns of lithic material selection also changed very little, as expected, if a model of cultural continuity is correct. Although coastal site density seems to have increased during the Ceramic period, there is no evidence supporting increased utilization of the PB lithic resource base. The presence of lithic resources may have been important in decisions of settlement location, but local resources rarely provided the bulk of all selected materials. In all but one assemblage, 50 percent or less of the materials came from sources near each site; 40 to 50 percent of the rocks originated from areas outside of the Bay. That people depended so much on imported materials is not indicative of their inability to adapt to the Bay’s lithic resource base. Importation may have been an adaptive response to environmental limitations, accomplished, as it had been earlier, through direct and/or indirect procurement of distant materials. Intersite variations in the levels of local PB, distant PB, and foreign rocks demonstrates the availability of several options for acquiring raw materials. Resource exploitation decisions often depended less on distance from the settlement than on tool 107 manufacturing or use requirements. These choices transcended time, space, and breaks in the cultural record. In conclusion, the people of Passamaquoddy Bay did not select a place to live with the sole objective of reducing time and energy in the exploitation of all resources. As they have for thousands of years, inventions of energy-efficient transport and exchange allow people to acquire items important to their existence that are lacking in areas where they live. Prehistoric people should be granted the same capacity for flexibility in dealing with their surroundings as held by those who study their cultural remains. 108 BIBLIOGRAPHY Ahler, Stanley A. 1975 Pattern and variety in Extended Coalescent lithic technology. Ph.D. dissertation, University of Missouri. University Microfilms, Ann Arbor. 1977 Lithic resource utilization patterns in the Middle Missouri subarea. In Trends in Middle Missouri prehistory, edited by W. Raymond Wood. Plains Anthropologist Memoir 13:132-150. Alcock, F. J. and G. S. Mackenzie 1960 St. $tephen, Charlotte County, New Brunswick. Geological Survey of Canada Map 1096A. Alcock, F. J. and S. C. Perry 1960 St. George, Charlotte County, New Brunswick. Geological Survey of Canada Map 1094A. Bailey, L. W. and G. F. Matthew 1872 Preliminary report on the geology of southern New Brunswick. Geological Survey of Canada, Report of Progress 1870-71, Part II. Baird, S. F. 1881 Notes on certain aboriginal shell mounds on the coast of New Brunswick and of New England. Proceedings of the United States Museum 4:292-297. Biard, Pierre 1897 Relation of New France, of its lands, nature of the country, and of its inhabitants. In The Jesuit relations and allied documents 3:Acadia 1611-1616, edited by Reuben G. Thwaites. Pageant Books, New York. Bishop, Jennifer 1980 Phil’s Beach: an artifact analysis and comparative study. Unpublished undergraduate thesis, Department of Anthropology, University of New Brunswick, Frederiction. Bonnichsen, Robson 1977 Investigating lithic sources for stone tool manufacture. Bulletin, Maine Archaeological Society 17:11-14. Bonnichsen, Robson and David Sanger 1977 Integrating faunal analysis. Canadian Journal of Archaeology 1:109-133. 109 Borstel, Christopher L. 1982 Archaeological investigations at the Young site, Alton, Maine. Occasional Publications in Maine Archaeology 2. Maine Historic Preservation Commission, Augusta. Bourque, Bruce J. 1971 Prehistory of the central Maine coast. Unpublished Ph.D. dissertation. Department of Anthropology, Harvard University. Bourque, Bruce J. and Steven L. Cox 1981 Maine state museum investigation of the Goddard site, 1979. Man in the Northeast 22:3-27. Brennan, Louis A. 1977 The.midden is the message. Archaeology of Eastern North America 5:122-137. Byers, Douglas S. 1959 The eastern Archaic: some problems and hypotheses. American Antiquity 24:233-256. Connolly, John 1977 Bear River, Nova Scotia: a collection analysis. Man in the Northeast 14:35-47. Cumming, L. M. 1967 Geology of the Passamaquoddy Bay region, Charlotte County, New Brunswick. Geological Survey of Canada Paper 65-69. Davis, Stephen A. 1978 Teacher’s Cove. New Brunswick Archaeology Series 1, 1. Historical Resources Administration, Fredericton. 1982 A Late Archaic cache/burial from New Brunswick. Man in the Northeast 24:135-146. Davis, Stephen A. and David Christianson 1979 Archaeological Resources in the Maritimes: a coastal survey. Report in Archaeology 4. Archaeology Branch, Historical Resources Administration. Fredericton, New Brunswick. D’Entremont, Stanley A. and Marilyn J. Moore 1977 An inventory of stemmed projectile points of Nova Scotia. Occasional Papers in Anthropology 3. Department of Anthropology, Saint Mary’s University, Halifax. Dincauze, Dena F. 1972 The Atlantic phase: a Late Archaic culture in Massachusetts. Man in the Northeast 4:40-61. 110 Erikson, Vincent 0. 1978 Maliseet-Passamaquoddy. In Handbook of North American Indians (Vol. 15-Northeast). Smithsonian Institution, Washington. Fitzhugh, William W. 1972 Environmental archeology and cultural systems in Hamilton Inlet, Labrador. Smithsonian Contributions to Anthropology 16. Washington. Fowler, William S. 1966 Cache of engraved pebbles from New Brunswick. Bulletin of the Massachusetts Archaeological Society 28:15-17. Funk, Robert E. and Bruce E. Rippeteau 1977 Adaptation, continuity and change in upper Susquehanna prehistory. Occasional Publications in Northeastern Anthropology 3. Man in the Northeast, Amherst, Massachusetts. Gadd, N. R. 1973 Quaternary geology of southwest New Brunswick with particular reference to Fredericton area. Geological Survey of Canada Paper 71-34. Ganong, William F. 1899 A monograph of historic sites in the Province of New Brunswick. Transactions of the Royal Society of Canada 105:213-357. Gates, Olcott 1967 Lower Silurian-Lower Devonian volcanic rocks of New England coast and southern New Brunswick. In North Atlantic geology and continental drift, edited by Marshall Kay. American Association of Petroleum Geologists Memoir 12:464-503. Gesner, Abraham 1839 First report on the geological survey of the Province of New Brunswick. Saint John. Gould, Richard A. 1980 Living archeology. Cambridge University Press. Gramly, Richard M. 1980 Raw material source areas and ’’curated” tool assemblages. American Antiquity 45:823-833. Grant, D. R. 1970 Recent coastal submergence of the Maritime Provinces, Canada. Canadian Journal of Earth Sciences 7:676-689. Hadlock, Wendell S. 1939 The Taft’s Point shell mound at West Gouldsboro, Maine. The Robert Abbe Museum, Bulletin 5. Bar Harbor, Maine. Ill Hay, P. W. 1967 Sedimentary and volcanic rocks of the St. Andrews-St. George Area, Charlotte County, New Brunswick. New Brunswick Department of Natural Resources, Mineral Resources Branch, Map Series 67-1. Jochim, Michael A. 1976 Hunter-gatherer subsistence and settlement: a predictive model. Academic Press, New York. Kingsbury, Isaac W. and Wendell S. Hadlock 1951 An early occupation site, Eastport, Maine. Bulletin of the Massachusetts Archaeological Society 12:22-26. Lahti, Eric, Arthur Spiess, Mark Heddon, Robert Bradley and Alaric 1981 Faulkner Test excavations at the Hodgdon site. Man in the Northeast 21:19-36. LaPierre, Paul 1974 Lithological classification scheme for the Hirundo, Maine and Passamaquoddy Bay, New Brunswick archaeological sites. Ms. on file, Department of Geology, University of Maine, Orono. Lavoie, Jacques 1972 Le site Sandy Point (BgDs:6) et 1’origin de la tradition sylvicole dans la region de la Baie de Passamaquoddy Noveau-Brunswick. Unpublished M.A. thesis, Department of Anthropology, University of Montreal. MacDonald, George F, 1968 Debert: a palaeo-Indian site in central Nova Scotia. National Museum of Canada Anthropology Paper 16, 011awa. Matthew, George F. 1884 Discoveries at a village of the Stone Age at Bocabec, N.B. Bulletin of the Natural History Society of New Brunswick 3:6-30. 1900 A quarry and workshop of the STone Age in New Brunswick. Proceedings and Transactions of the Royal Society of Canada 6:61-69. McCormick, James S. 1980 Criteria in archaeological faunal analysis: four Passamaquoddy Bay cases. Unpublished M.S. thesis, Institute for Quaternary Studies, University of Maine, Orono. McGuire, Joseph D. 1908 Ethnological and archeological notes on Moosehead Lake, Maine. American Anthropologist 10:549-557. 112 Moffett, Rose 1952 Letter to Isaac W. Kingsbury, from Provincehaven, Mass. Letter housed at the R. S. Peabody Museum, Andover Massachusetts. Munsell Color Company, Inc. 1971 Munsell Soil Color Charts. Munsell Color Company, Inc. Baltimore. Osberg, Philip H., Arthur M. Hussey II and Gary M. Boone 1984 Bedrock geologic map of Maine. Department of Conservation, Maine Geological Survey Map No. 84-1. Osborne, Douglas 1965 Chipping remains as an indication of cultural change at Wetherill Mesa. In Contributions of the Wetherill Mesa archaeological project. Memoirs of the Society for American Archaeology 19:30-44. Pavlides, Louis 1962 Geology and manganese deposits of the Maple and Hovey Mountains area, Aroostook County, Maine. U.S. Geological Survey Professional Paper 362, Washington. Pearson, Richard 1970 Archaeological investigations in the St. Andrews area, New Brunswick. Anthropologica 12:181-190. Potter, R. R., E. V. Jackson and J. L. Davies 1968 Geological map, New Brunswick. Department of Natural Resources, New Brunswick Map N.R.-1. Reher, Charles A. and George C. Frison 1980 The Vore site, 48CK302, a stratified buffalo jump in the Wyoming Black Hills. Plains Anthropologist, Memoir 16. Renfrew, Colin 1977 Alternative models for exchange and spatial distribution. In Exchange systems in prehistory, edited by Timothy K. Earle and Jonathan E. Ericson, pp. 71-90. Academic Press, New York. Rhoades, David A. 1963 The geology of the Perry Formation. Unpublished M.S. thesis, Department of Geology, University of Maine, Orono. Ritchie, William A. 1961 A typology and nomenclature for New York projectile points. New York State Museum and Science Service Bulletin 384, Albany. 113 Ritchie, William A. 1969a The archaeology of Martha’s Vineyard: a framework for the prehistory of southern New England. The Natural History Press, Garden City, New York. 1969b The archaeology of New York State (revised edition). The Natural History Press, Garden City, New York. Ruitenberg, A. A. 1968 Geology and mineral deposits, Passamaquoddy Bay area. New Brunswick Department of Natural Resources, Mineral Resources Branch, Report of Investigation 7. Ruitenberg, A. A. and Allan Ludman 1978 Stratigraphy and tectonic setting of early Paleozoic sedimentary rocks of the Wirral-Big Lake are, southwestern New Brunswick and southeastern Maine. Canadian Journal of Earth Sciences 15:22-32. Sabina, Ann P. 1972 Rock and mineral collecting in Canada: New Brunswick, Nova Scotia, Prince Edward Island and Newfoundland (Vol. III). Revised edition. Geological Survey of Canada Miscellaneous Reports 8. Ottawa. Salmon, Merrilee H. 1978 What can systems theory do for archaeology? American Antiquity 43:174-183. Sanger, David 1971 Prehistory of Passamaquoddy Bay: a summary. Bulletin, Maine Archaeological Society 2:14-19. 1973 The prehistoric components on St. Croix Island, Maine. Ms. on file, Department of Anthropology, University of Maine, Orono. 1975a The Pagan Point archaeological site in St. Andrews, New Brunswick: preliminary report. Ms. on file, Department of Anthropology, University of Maine, Orono. 1975b Draft manuscript of Passamaquoddy Bay prehistory, personal file of D. Sanger 1976 The earliest settlements: 9000 B.C. to A.D. 1600. In Maine forms of American architecture, edited by Deborah Thompson, pp. 3-14. Colby College Museum of Art, Camden, Maine. 1979 The Ceramic period in Maine. In Discovering Maine’s archaeological heritage, edited by D. Sanger, pp. 99-115. Maine Historic Preservation Commission, Augusta. 114 Sanger, David 1980 Archaeological salvage and test excavations, Fernaid Point, Acadia National Park. Report to the National Park Service, Department of the Interior. 1981a Archaeological investigations at Frazer Point and Duck Harbor, Acadia National Park, Maine. Report to the National Park Service, Department of the Interior. 1981b Unscrambling messages in the midden. Archaeology of Eastern North America 9:37-42. 1982 Changing views of aboriginal seasonality and settlement in the Gulf of Maine. Canadian Journal of Anthropology 2:195-203. n.d. The Carson site, a shell midden in Passamaquoddy Bay. Manuscript in preparation. Sanger, David, Ronald B. Davis, Robert G. MacKay and Harold W. Borns, 1977 Jr. The Hirundo archaeological project—an interdisciplinary approach to central Maine prehistory. In Amerinds and their paleoenvironments in northeastern North America, edited by W. S. Newman and B. Salwen. Annals of the New York Academy of Sciences 288:457-471. Sanger, Mary E. J. 1968 Preliminary report of the 1968 excavation at the Eidlitz site (BgDs:4), St. Andrews, New Brunswick. Ms. on file, Historic Preservation Commission, New Brunswick. Struever, Stuart 1971 Comments on archaeological data requirements and research strategy. American Antiquity 36:9-19. Thomas, Martin L. H. 1983 Marine and coastal systems of the Quoddy region, New Brunswick. Canadian Special Publications of Fisheries and Aquatic Sciences 64. Department of Fisheries and Oceans, Ottawa. Willoughby, Charles C. 1901 Prehistoric workshops at Mt. Kineo, Maine. The American Naturalist 35:213-219. 115 APPENDIX Recommendations Any archaeologist contemplating research of prehistoric lithic resource exploitation might consider the following observations made in and recommendations resulting from this study. Where appropriate, applications to nonlithic resource exploitation studies can be made. 1. Observation: Among sites in a settlement area, utilization of different lithic resources was affected by the distribution of those resources. Example: Differences in selection for quartz and for other indigenous materials. Recommendation: The researcher should be aware that distributions of nonlithic resources may have also affected their material distributions through the sites. 2. Observation: Different rocks were used for making different kinds of tools. Example: Differences in material choices for unifaces and bifaces. Recommendation: Assemblages should be divided according to tool classes before comparing lithics in collections from different sites. 3. Observation: Selection patterns may differ between components within one site. Example: Differences in material selection between Aceramic and Ceramic stemmed bifaces in the Minister’s Island (BgDslO) and Teacher’s Cove (BgDrll) sites. Recommendation: Researchers should separate tools from different periods before attempting comparisons of raw material selection patterns. 116 4. Observation: Selection for material texture may have changed even if selection for rock type did not. Example: Differences in volcanic textures between Aceramic and Ceramic stemmed bifaces. Recommendation: In addition to lithic material types, material texture should be recorded. 5. Observation: People living in neighboring sites did not make equal use of the same resources. Example: Different selection patterns among the eastern PB sites. Recommendation: Collections from more than one site in a settlement location should be compared for intersite variability. 6. Observation: Groups of sites in different sectors of the settlement area had different lithic resource utilization patterns. Example: Difference between sites of eastern PB and western PB sectors. Recommendation: Site selection should accommodate the spatial distribution of indigenous resources. 7. Observation: Patterns observed in a settlement area may not be typical of patterns in other areas of the cultural region. Example: Difference observed in material choices for unifaces between PB sites and synchronous sites in Maine. Recommendation: The researcher should not base regional exploitation patterns on those observed in one settlement area. Final Recommendations The spatial and temporal boundaries of material selection patterns observed in the Passamaquoddy Bay sites needs to be ascertained. Comparisons with sites in nearby areas of equal or older time periods would demonstrate whether some patterns were idiosyncratic to Passamaquoddy Bay. 117 More research of Aceramic patterns of subsistence and settlement is needed to discover if other behaviors besides *any governing lithic resource exploitation remained constant as a way of life was transformed 2000 years ago. 118 BIOGRAPHY Anita Lynne Crotts was born in Washington, D.C. on April 28, 1950. She received her early education in public schools in Washington, D.C. and Springfield, Virginia, and was graduated from Annandale High School in 1968. She was enrolled in George Mason College of the University of Virginia and Catholic University from September 1968 to June 1971, and received a Bachelor of Arts degree in Anthropology (special honors) from George Washington University in June 1972. At this time she began employment with Meloy Laboratories, Inc., in Springfield, Virginia. In September, 1975, she was enrolled for graduate study at the University of Maine and served as a graduate assistant in the Institute for Quaternary Studies, Anthropology Department. During this time, she was employed in cultural resource management by several Maine state agencies and the United States Forest Service, Asheville, North Carolina. She is a candidate for the Master of Science degree in Quaternary Studies from the University of Maine at Orono in December 1984.