Prehistoric Fishweirs in Eastern North America

Master's Thesis by allen lutins


Submitted in partial fulfillment of the requirements for the degree of Master of Arts in Anthropology in the Graduate School of the State University of New York at Binghamton, May 1992

Accepted in partial fulfillment of the requirements for the degree of Master of Arts in Anthropology in the Graduate School of the State University of New York at Binghamton:
Charles Cobb, Department of Anthropology {signature}
Randall McGuire, Department of Anthropology {signature}

Copyright © 1992-2014 by allen lutins.

ABSTRACT

Fish weirs, or semi-permanent traps aimed at the exploitation of aquatic resources, occur throughout the eastern seaboard of North America. Many, perhaps hundreds, are of prehistoric construction. Their existence is virtually unrecognized in the archaeological literature for eastern North America, potentially resulting in inadequate reconstructions of subsistence and settlement patterns in this region. This thesis is an attempt to synthesize the information for all known prehistoric weirs in eastern North America, and to analyze that information for its importance in reconstructing prehistoric subsistence and settlement patterns.

Note

This is the text of my thesis as it was written in 1992. Since that time, a few strides have been made in fish weir research, particularly the new research at Sebasticook Lake (Maine), renewed research at Boylston Street (Boston, Massachusetts), and most significantly the upcoming publication of John Connaway's
Fishweirs - A World Perspective. There's one HTML feature which may not be obvious: If you click on a note number, which looks like this: [3] you'll be put at the "Notes" section at the end; by clicking on the Note number (which looks like 3.) you'll be returned to your place in the text.

Contents

INTRODUCTION REASONS FOR PAUCITY OF DATA ON WEIRS AND FISHING GENERAL DISCUSSION OF WEIRS KNOWN PREHISTORIC WEIRS IN EASTERN NORTH AMERICA ANALYSIS OF WEIR USE IMPLICATIONS OF WEIR USE

Acknowledgements

This project would have been entirely impossible if not for leads and tidbits of information passed on to me by numerous people over the years. Among those who were particularly helpful were (in alphabetical order) John Cavallo, Dena Dincauze, Jonathan Gell and the rest of the staff at New Jersey DEP's Office of New Jersey Heritage (as well as the staff of other DEP offices, including the Division of Coastal Resources and the Division of Fish, Game and Wildlife), Rob Jacoby, Bill Oliver, Leonid Shmookler, and Vin Steponaitis. Bill Frazier is to be especially commended for the generous amount of information and photocopies which he provided me; and I thank the staff of the Inter-Library Loan office at the Bartle Library at SUNY-Binghamton for processing my numerous (and often arcane) requests.

Special thanks go to Randy McGuire, who provided input and valuable feedback on previous incarnations of this project, and both Randy and Charlie Cobb for their editorial advice and guidance. Charlie was particularly helpful in helping me to locate sources of information pertaining to southeastern archaeology.

For minor but necessary contributions, I gratefully acknowledge the assistance of the New Jersey State Museum staff, and the various archaeologists who have served in the Army Corps of Engineers, New York District over the years.

Last but not least, I owe perhaps my greatest debt to my fellow graduate students, too numerous to name, who not only provided me with tips on sources and other useful information, but also maintained the creative, supportive, and cooperative environment necessary for my completing such a task.

Any omissions, oversights, or misrepresentations in this paper are completely my own, and do not reflect deficiencies on the part of the above-named people.

Dedication

This thesis is dedicated to Dave Wasilewski, who more than anyone else, helped me to keep my sanity during my years in Binghamton. When I need a break, Dave is always there; and when I need to get out of the area, Dave is always a willing partner in road trips, back-packing treks, and the like. The few times he wasn't available, Dave generously allowed me the use of his car so that I could escape on my own.

Introduction

Importance of fishing in eastern North America

Archaeologists concentrating on eastern North America have only recently begun to realize the importance of fishing to the reconstruction of prehistoric subsistence patterns in eastern North America (Brumbach 1986:6, 35-36,62; Wheeler and Jones 1989:1). As a result, little is known about prehistoric fishing practices in much of the region (e.g., Rostlund 1952:70; Kraft 1984:5). Relatively little work has been done in the region which directly addresses this specific topic (Brumbach 1978:225). In addition, a general scarcity of data pointing to the importance of fish in prehistoric diets contributes to this lack of knowledge. Even eastern North America, the settlements, structures, and other accoutrements associated with fishing are, in general, poorly known (Moss and Erlandson 1990:143).

It is likely that fish played an important role in the diet of peoples living in the vicinity of lakes, rivers, and the Atlantic coast, as this was the case for peoples the world over beginning in the Upper Pleistocene (Wheeler and Jones 1989:xii). Fish are an attractive food source because little risk and effort is expended in catching them (Perlman 1980:260,277). Fish are highly fecund, and are able to repopulate quickly despite heavy predation (Wheeler and Jones 1989:4). Among hunter-gatherer/collectors, the only alternatives providing such high yields per effort expended are deer and nuts (Perlman 1980:281). In northeastern North America, Funk (1983) and Snow (1980) claim that only the former were utilized for periods pre-dating the Early to Middle Woodland based on an absence of tools commonly associated with nut-meat extraction.

Numerous contact-era accounts attest to the widespread availability and extensive exploitation of fish throughout the entire Atlantic seaboard (Brumbach 1986:37) [1]. Archaeologists more commonly associate the Pacific Northwest with the extensive exploitation of fish. However, the Atlantic was capable of providing an equal abundance of this resource, with approximately one hundred native species large enough to warrant use as a food source (Brumbach 1986:41; Rostlund 1952:7).

Fisher (1983:36) characterizes anadromous [2] fish as constituting "probably the greatest density of available food that existed for pre-agricultural societies in the Hudson Valley." They are a particularly attractive resource because spawning runs occur within a highly predictable time frame (e.g., Fisher 1983:40; Kraft and Mounier 1982:61; Funk 1983:337). These runs can usually be accurately predicted to within a few days (Schalk 1977:213). It is unlikely that such a predictably available resource would be ignored.

Fishing and the archaeological record

"Fishweirs," or simply "weirs," are traps built in coastal, estuarine or riverine waters to capture large numbers of fish (Schalk 1977:232). Although they are represented in the archaeological record throughout eastern North America, there has been an absence of coordinated research on them. Despite destruction due to age or from urbanization, a number of weirs still survive, as do numerous references (both early ethnohistoric descriptions of their use and more contemporary descriptions of their presence) to weirs no longer extant.

In some parts of the world, prehistoric weirs have received considerable study [3]. Only one source, Rostlund's Freshwater Fish and Fishing in Native North America (1952), discusses eastern North American weirs somewhat comprehensively. Despite listing dozens of areas containing weirs, Rostlund presents no map of weir locations nor accurate sketches of weirs. It is clear that Rostlund omitted a number of sources of information which were available to him. Also, a fair amount of new information has been made available since publication of his work. Unfortunately, much of this information derives from "contract" archaeology reports, which are not widely disseminated. This led Johnston and Cassavoy to conclude (1978:708): "An earnest search of the accessible literature on weirs has revealed that it leaves much to be desired and, with one exception, that there is a virtual absence of useful references to archaeological weirs."

The paucity of references devoted to fishweirs is not an indication of their rarity. Peck (1977:5) claims that "(t)housands of stone fish traps occur throughout the Piedmont region along the Atlantic Seaboard." This is likely an exaggeration, even including historic and modern weirs. To date, weirs in only four locations have been adequately sketched and accurately mapped [4]. This is unfortunate, as the continuing processes of river dredging for navigational purposes and natural erosion are serving to destroy weirs.

Archaeologists associate a number of general assumptions with weirs. One is that weirs are usually associated with the exploitation of anadromous and catadromous fish (Schalk 1977:232). Another is that they are temporally and spatially non-diagnostic (Flannery 1939:17). Given the paucity of research undertaken regarding weirs, it is clear that such hypotheses are often untested, especially regarding weirs in eastern North America.

The inadequacy of information regarding fishweirs also impinges on reconstructions of prehistoric subsistence and settlement patterns. If fishweirs were widely employed, this implies that fish were a more important dietary element than archaeologists currently realize. Similarly, the use of weirs may have affected settlement patterns because of the time and seasonal constraints necessitated by their construction, repair, and utilization.

I will summarize the available literature on prehistoric weirs in eastern North America, and analyze their construction and use. A discussion of related issues (such as consequences for the reconstruction of subsistence and settlement patterns) follows, in an attempt to better understand the role of fishing in prehistoric eastern North America, and the nature of weirs themselves. Particular attention will be paid to the possibilities of determining the spatial and temporal variability of weirs, and to augmenting our general knowledge of prehistoric subsistence and settlement patterns, based on the study of weirs.

Reasons for paucity of data on weirs and fishing

Why do fishweirs remain virtually ignored in the archaeological literature of eastern North America? One can point to three basic deficiencies as the cause: A lack of archaeological data, a shortage of historical references, and a paucity of contemporary research on the subject of weirs.

Lack of archaeological data

Archaeological data concerning the role of fish weirs and fishing are lacking for a number of reasons. These can be broken down into three basic categories: the nature of fishing sites and their associated features, lack of preservation, and sampling biases. Perhaps we can include in the latter a relative shortage of financial support given to large-scale and/or long-term projects in the northeastern United States, as compared with other parts of the country (Kraft 1984).

Nature of fishing sites and weirs

Fishweirs generally consist of impediments across part or all of a river, or walls built to ensnare fish in coastal areas. Most surviving weirs are of stone, rendering direct dating impossible (Godwin 1988:52; Johnston and Cassavoy 1978: 708). This means that most archaeological dating of weirs derives either from direct dating of extant stake weirs (of which there are few), or indirect dating based on associated materials (Godwin 1988:53).

Aside from the weirs themselves, there is a general lack of material technology associated with the exploitation of fish (Whyte 1988:115; Brumbach 1978:7). Weir-fishing renders the use of recognizable fishing implements (such as hooks, net sinkers, etc.) unnecessary. This may account (in part) for the lack of such tools in the archaeological record overall (Fisher 1983:37). It is also likely that some fish-processing tools and features go unrecognized by archaeologists (Brumbach 1986:37). For example, Fisher (1983:39) reports that in three seasons of excavations at the Shantok Cove site in Connecticut, only one fishing implement (which he fails to identify) was found, yet thousands of fish bones were recovered. No artifacts typically associated with fishing activities were found during limited excavations in the vicinity of known weirs in the Potomac (Strandberg and Tomlinson 1969:319).

Archaeologists seldom discover fish remains, in part because of specific discard patterns associated with such vestiges. Ethnographic and ethnoarchaeological reports indicate that people usually discard (and process) fish away from habitation areas, or else they thoroughly burn fish remains, because of their strong smell and propensity to attract scavengers (Brumbach 1986:37,44,46; Whyte 1988:115). In general, fish bones are relatively small in size, further hampering efforts aimed at their discovery. Among some groups, fish bones are valued as a food source during periods of scarcity, so they may be consumed along with the meat and broken down in the intestinal tract. The bones that remain are often destroyed in the intestinal tracts of scavenging animals (Brumbach 1978:222; Jones 1986:53).

Preservation

Problems with preservation in the archaeological record hinder our knowledge of prehistoric fishing. Stake weirs are preserved only under extraordinary conditions (Johnston and Cassavoy 1978:708). Acidic soils, prevalent in the Northeast, tend to destroy osteological remains, especially the small, delicate bones of fish (Brumbach 1978:7; Brumbach 1986:37). Under some conditions, fish bones may be preserved, but the techniques and expertise required for their recognition, recovery and analysis are currently inadequate (Wheeler and Jones 1989:3-7).

The destruction of entire sites along the Atlantic seaboard further hampers our ability to understand the importance of fishing in this region. Due to sea level rises previous to 3,000 to 4,000 B.P. the entire eastern seaboard has been inundated. As a result, prime locations for fishing (i.e., coastal areas, lagoons, and estuaries) dating to before the Woodland period are now typically 50 to 60 meters below sea level (Emery and Edwards 1966:735). In addition, those surviving sections of the eastern seaboard which were ideal for fishing were focal points for early European settlement, and are now virtually inaccessible to archaeologists due to intensive urban development (Kraft 1984:3).

Sampling biases

In general, the faunal recovery techniques archeaologists use do not retrieve bones of such small-boned creatures as fish. Better methods of recovery, such as flotation and water screening, have been known to archaeologists in eastern North America for some time now; however, their adoption to date in this area is not widespread (Brumbach 1986:37; Wheeler and Jones 1989:7).

Lack of historic data

Quite a few ethnohistoric accounts depicting native fishing at the time of European contact survive, and will be related herein. By the time Europeans reached eastern North America, fishing was likely in decline, owing to an increased dependence on agriculture (Brumbach 1978:6; Ritchie 1980:278). In addition, Brumbach (1986:36) notes that "agricultural activities were given undue emphasis, perhaps, in the observations and descriptions of the early European colonists." Therefore, fishing was likely more important throughout prehistory than is suggested by these accounts.

Contemporary folklore and stereotypes that we are exposed to contribute to a lack of knowledge concerning native American fishing practices. Brumbach (1986:36) noted that "popular folklore emphasizes fertilizer value of the fish but seems vague about their consumption as food." Perhaps the stereotype of the "hunter/gatherer" among anthropologists similarly attenuated a focus on fishing, as the word "fishing" is not included in the phrase "hunting/gathering." Despite this fact, in some societies, the role of fishing may have been equal to or surpassed that of hunting and/or gathering.[5]

Lack of contemporary research

The depletion of east coast fisheries is associated with overexploitation by early European settlers. This renders contemporary reconstructions of prehistoric fishing conditions difficult to impossible (Brumbach 1978:225; Brumbach 1986:50). The depletion of fisheries is a problem with roots centuries old (Brumbach 1978:250; Brumbach 1986:61). Concerning early colonial New England, Bakeless (1961:223) reports:
Plymouth set the (fishing) limit between five hundred and one thousand barrels (of fish) for the town as a whole. By the Revolution this had come down to two hundred barrels. In 1730 each household was limited to four barrels. In 1763, the towns of Plymouth and Wareham took 150 barrels from Agawam Brook alone. Sad to say, in the Merrimac River, salmon, shad, and alewives were growing less abundant as early as 1753.
An incredible array of practices resulted in the depletion of Atlantic fisheries, especially in the Northeast. These included water diversion for canals, thermal and chemical pollution, dam-building, dredging for navigational purposes, and over-fishing (Brumbach 1978 228, 251-2; Brumbach 1986:61-2; Brydon 1974). In addition, a number of species were displaced (sometimes intentionally, sometimes accidentally) by non-native species such as the carp and rainbow trout (Brumbach 1986:36; Wheeler and Jones 1989:34).

Weirs themselves suffer from a lack of research in eastern North America. Both historic and prehistoric weirs dot the landscape, yet they seem to have escaped the attention of historians and archaeologists alike. Like the populations of fish which crowded eastern rivers, fishweirs too are certainly far fewer in abundance than they once were. Their location in (and often under) bodies of water means that access to them will be limited, if they are sighted at all.

General discussion of weirs

Definition

Archaeologists often draw a distinction between weirs and traps (see, for example, Chapman 1975:3). Erhard Rostlund, in a comprehensive discussion of weirs in North America, notes that "the function of every weir is to obstruct the passage of the fish in order to facilitate its capture, and the purpose of every trap is to impound the fish so that it cannot get away" (Rostlund 1952:101). Johnston and Cassavoy (1978:706) reiterate this distinction, but note: "Weirs and traps may be used independently but their functions are often combined in a single structure when traps are used to collect the fish forced by the weir to a narrow outlet." Thus, it is not always a simple matter to distinguish between the two. In this thesis, any large-scale, relatively permanent
[6] device used to aid in the capture of large numbers of fish will be referred to as a "weir." Also, the term "tidal weir" will be used refer to any device (again, large-scale and relatively permanent) used to trap fish "behind an obstruction with the fall of the tide," following Rostlund's definition of a "tidal trap" (Rostlund 1952:101).

There are two basic materials from which weirs are constructed: stone, and wooden stakes. Archaeologists sometimes refer to "brush" weirs, but the term `stake' will be used throughout this thesis to distinguish a device made from upright poles. Stake weirs are generally constructed of similar-sized poles, as opposed to `brush,' which is interpreted herein as the placement of piles of branches without regard to size and length (this latter method is sometimes employed as part of a weir, as will be seen later).

Location

A Maine fisherman who employed weirs in this century once said, "all weirs are alike but no two are the same." This refers to the fact that all weirs were used for the same purpose, but each is especially adapted to its particular environmental and topographic setting (Johnston and Cassavoy 1978:705). In order to be practical, weirs must be built in locations where fish are in abundance for at least part of the year (Johnston and Cassavoy 1978:705). If they are not moving in large numbers of their own accord (i.e., during spawning runs), the fish must be induced to do so (Rostlund 1952:101).

Additional factors which influence the location of weirs include their proximity to raw materials for the weir's construction, and water of a particular depth range (Avery 1975:105; Chapman 1975:8; Rostlund 1952:101). For tidal weirs, this may mean a location where water levels vary only 2.2 meters between high and low tides (Avery 1975:111) [7]. Riverine weirs commonly seem to be built on existing shoals and in other naturally shallow areas.

The sandy nature of sea bottoms and the force of ocean gales and waves on the coast add to the complications of fishweir construction. Tidal weirs were likely restricted to lagoons and estuaries to avoid some of these difficulties (Larson 1980:10-11).

The construction of riverine traps carries different concerns. They must be built in rivers that are not so swift as to destroy the weirs, yet are not too sluggish (Chapman 1975:8; Brinkhuizen and Clason 1983:15). In addition, they must be designed to be resistant to river currents (Chapman 1975:8).

Construction

Technique

Based on observations from Alaska, Australia, Japan, and southern Africa, weirs seem to conform to one of three basic designs: The first is the tidal weir. The second is a "maze-like" arrangement of walls; the design is such that fish easily find their way in, but can not easily escape. This design is usually used in estuarine areas, and so may be defined as a type of tidal weir. The third is an obstructing wall which funnels fish to a particular point at which they may be trapped or removed (Godwin 1988:52). This is the form used in all riverine structures. There are perhaps other types of weirs which do not fall into these basic categories, but they are rare, and there are no reports of them from eastern North America
[8]. While conforming to these three basic designs, weirs are constructed with a great diversity of specific shapes and materials.

Both tidal (lagoon and estuarine) and riverine weirs are subject to damage from the elements, including excessive currents (flooding in riverine/estuarine environments and gales in lagoon/coastal environments) and damage from large floating objects. Therefore, it is likely that weirs needed to be rebuilt from time to time (Johnston and Cassavoy 1978:706-7).

Difficulty

How difficult is it to build and maintain weirs? It is possible that some weirs are relatively simple to build. However, this view is based on analogies with simple contemporary projects or with weirs from other parts of the world, rather than the large stone or stake weirs known from eastern North America (for example, Godwin 1988:56; Wheeler and Jones 1989:28,171). Swezey and Heizer (1984:973) make it clear that it often takes a relatively large number of people in order for weir construction to be an easy task. Alternatively, weir construction may be relatively labor-intensive (Hubbert and Wright 1987:98; Schalk 1977:232). This assumption is similarly flawed, apparently being based on simple observations of stone weirs, rather than controlled studies or observations of their construction (e.g., Peck 1977:3; Larson 1980:123). In truth, it is almost certain that the difficulty of weir building depends on a number of factors, including the number of people available, the topography of the location, the availability of materials, and the design of the weir.

Use

Method

Descriptions of weir use in the archaeological literature are extremely contradictory. I will relate the basic methods of weir use as reconstructed in the archaeological literature. Actual descriptions of weir use from ethnographic accounts will follow, succeeded by discussions relating to these reconstructions and descriptions.

Weirs can be used in either an active or passive manner, or occasionally a combination of the two (Brinkhuizen and Clason 1983:15). Tidal weirs are an example of the latter; they simply trap fish behind an obstruction with the fall of the tide. One then collects the fish at a convenient time. Riverine and estuarine weirs (and perhaps some coastal weirs as well) may be used either way; fish may be ensnared in large numbers of their own accord during spawning runs, or they may be corralled in the weir through human intervention (Rostlund 1952:101). All of the descriptions of weir use in the ethnographic accounts to follow fit these patterns.

Species selectivity

As with discussions of weir construction and use, there is no consensus concerning the species selectivity of weirs. Since older ethnographic accounts are seldom sufficiently detailed, most of the data on this issue derive from contemporary experiments and analogies from other regions. Some reports claim that the variety of species obtained in a weir is quite small, but these generally derive from analogies with fishing in Africa, Tasmania, or other regions whose comparison to the situation in eastern North America is questionable. Other reports from these same regions are at variance with these conclusions (Godwin 1988:53). A similar debate exists as to the size range of fish caught with weirs (Godwin 1988:54; Wheeler and Jones 1989:28). Literature discussing the species and size selectivity of fish trapped in weirs is insufficient from which to draw definitive conclusions, but experimental evidence suggests that, with little difficulty, traps can be constructed so as to select for particular size ranges of fish (Mahaffy 1978:59).

Seasonality

Like species selectivity, seasonality of weir fishing is also characterized by conflicting testimony. Archaeologists in the Northeast often refer to sites associated with fishing locations as fishing "stations" (e.g., Snow 1980:200,216,230). They often assume that such sites were occupied for short terms only in both the Northeast (e.g., Kraft and Mounier 1982:61; Fisher 1983:36; Lenik 1985:153; Williams and Thomas 1982:122) and the Southeast (e.g., Jones 1873:324; Jenkins and Krause 1986:41,119). However, assumptions of short term usage are unjustified. There are areas (particularly in the Southeast) where fish are available nearly year-round. Due to species diversity in the Southeast, anadromous fish runs take place throughout much of the year (Schalk 1977:216,241).

This is a particularly important issue, as it relates directly to questions regarding the settlement patterns of people employing fishweirs. Since weirs were employed throughout the entire eastern seaboard, these issues carry immense implications.

Known prehistoric weirs in eastern North America

The following discussion of prehistoric fish weirs is based on ethnographic accounts, historic and contemporary descriptions of extant weirs, and contemporary archaeological findings. They are presented primarily from north to south, starting with the western Great Lakes region, and proceeding thence to New England, then southward along the Atlantic coast.

It usually difficult to ascertain the age of a fishweir with any degree of certainty. The three basic criteria used in this thesis include contemporary archaeological analyses, contact era accounts, and historic (i.e., eighteenth, nineteenth, and early twentieth century) descriptions. Of these criteria, archaeological analyses which include modern dating techniques are the most reliable indicator of age. Unfortunately, few weirs have been so analyzed to date.

Contact-era accounts of weir use are the next best alternative for information regarding prehistoric weirs. Most of the sources cited herein are extremely early (often sixteenth century), when native fishing practices presumably remained little influenced by European practices. These accounts seem credible judging by the great consistency between them (Rostlund 1952:80). Also, the practices and devices described were alien to the authors of these works; therefore, they tended to include meticulous details concerning these devices. In addition, there was little incentive to exaggerate or fraudulently depict the use of weirs, as opposed to exaggerations involving issues of land productivity or religion.

Historic accounts dating to the last few centuries are less reliable sources of information concerning prehistoric weirs. Prehistoric weirs were sometimes refurbished and re-used by both Native Americans and colonists during historic times (Strandberg and Tomlinson 1969:313), and new weirs were constructed by European colonists. Therefore, when historic accounts claim that a particular weir or set of weirs is prehistoric, every attempt was made to discover corroborating or contradictory evidence for such claims.

This section is divided into sections based primarily on contemporary state boundaries. Although this is an admittedly arbitrary delineation (bearing in mind that these boundaries did not exist throughout the time period when these structures were built), there are two basic reasons for retaining this hierarchy: First, most of the overviews from which these data are derived consist of surveys done on a state-wide basis. Second, it will be important for comparative purposes to separate these accounts by region, and this method will permit a high degree of specificity in this regard.

Midwestern U.S.

Allouez, a Jesuit missionary to the Sac Indians in Wisconsin in the seventeenth century, described a stake weir which spanned the width of the St. Francis River, used to catch "Sturgeon and every other kind of fish" in the spring and summer (Brumbach 1986:39).

Only one extant weir has been described from the midwest, a stone structure located on the Chariton River (near its confluence with the Missouri River) in Missouri (Connaway 1982:156). Shields (1967:490) presents evidence that this was indeed a prehistoric structure, and that it was originally "V"-shaped (with a gap in the center), but was reconstructed (and its design altered) by European Americans sometime after 1837. The structure was repeatedly dynamited in the early twentieth century (after several sequences of rebuilding) by the Missouri Fish and Game Commission when weir fishing was outlawed (Shields 1967:491).

A postcard in this author's possession depicts a stone weir on the Iowa River (in Iowa) which is reputed to be prehistoric (Figure 1). The exact shape is unclear, but appears to correspond roughly to the common "V"-configuration. Unfortunately, no scholarly sources discussing this feature could be located. The text of the postcard indicates that this structure "can only be seen in times of low water."

St. Lawrence Drainage

Peter Kalm recorded the use of weirs in this region in his diary on 4 August 1749. He described a tidal weir on the St. Lawrence River near Quebec, employed predominantly for catching eels, "made of twisted oziers, so close that no fish can get through them...on one side near the bottom is an entrance for the fishes, made of twigs, and sometimes of yarn made into a net." (Benson 1964:423-424). LeJeune described these weirs at an earlier date, mentioning their use by the Montagnais (Brumbach 1986:40). Again, eels were mentioned as the predominant species obtained. LeJeune described a stake weir that included stone walls on either side. The earliest known reference to these weirs dates to the sixteenth century, when Samuel de Champlain noted: "(a) great catch of fish takes place by means of a number of weirs which almost close the strait, leaving only small openings where they set their nets in which the fish are caught..." (Johnston and Cassavoy 1978:698). Neither this description nor that of Peter Kalm mentions stone elements.

Archaeological investigations demonstrate that weirs were in use in the St. Lawrence drainage long before European contact. Between 1973 and 1974, an archaeological investigation at the outlet of Lake Simcoe, at Atherley Narrows, Ontario, revealed 535 stakes clearly arranged for use as fish weirs (Johnston and Cassavoy 1978). These were accompanied by a feature composed of rocks, presumably to serve as a footing in this area of soft mud (Johnston and Cassavoy 1978:704). In one area, the stakes ranged from one-half to three inches in diameter, and were oriented "to obstruct fish swimming with the current toward Lake Couchiching and direct them toward the outlet in the vicinity of the rock-covered bottom" (Johnston and Cassaovy 1978:704). In another area, Johnston and Cassavoy claim that the stakes are "oriented on the diagonal northwest-southeast and would serve to obstruct fish swimming upstream toward Lake Simcoe". Radiocarbon dates on a number of the stakes demonstrate that the weir was constructed during the Late Archaic (Johnston and Cassavoy 1978:704) [9].

An unidentified early colonist commented that the Micmac employed "a fence of wood clear across the river to hinder the passage of fish" in which an opening was left for a net trap (Rostlund 1952:170). This description almost certainly refers to a weir constructed of stakes or brush, but neither the author, a specific date, nor the river(s) in question are provided.

Wintemberg (1902) makes mention of stake alignments, which appear to be the remains of weirs, in Burgess Lake, near Drumbo, Ontario. The feature consists of a double row of stakes aligned in an "L"-shape at a point presumed to be the prehistoric shore of the lake (Wintemberg 1902:36).

Iroquois culture area

Contact-era accounts attest to the use of weirs by the various nations of the Iroquois Confederacy. Beauchamp (1900:133) mentions that the Onondagas and Oneidas employed eel weirs (of an indeterminate type) on the Onondaga River at Caughdenoy. He claims that "(e)arly travelers described these," although he provides no references. He may be referring to the account of Dablon, a Jesuit missionary to the Oneida, who wrote in 1670: "Our savages construct their dams and sluices so well, that they catch at the same time the Eels, that descend, and the Salmon, that always ascends" (Brumbach 1986:40). Again, a description of this weir is lacking. Jacques Bruyas, an Iroquois linguist, commented in the mid-19th century that the Iroquois of his time were using stone weirs lined with brush, with box traps set into them (Beauchamp 1905:148).

An 1894 Bureau of American Ethnology report noted at the time that "(s)tone fish weirs yet remain in some New York streams, though many have been destroyed." That report specifically mentioned a stone weir in the Seneca River, and also refers to an account dating nearly 100 years earlier attesting to the existence of stone weirs in the Seneca River. This earlier account noted that they were "V"-shaped, being "well made of field stones of considerable size" (Thomas 1894:549). In 1900, Beauchamp described one of the Seneca River weirs (reported to be near Baldwinsville) as having "three bays of unequal length reaching up to the river as it tended to the north shore. It was built of fieldstone and was about 1,200 feet long." He claimed that the remains of a second weir existed nearby, and that "others are found elsewhere" (Bradley 1987:210).

New England (non-specific)

Trigger (1978:153) maintains that weirs (of an unspecified type) were in use by the Western Abenaki at the time of European contact. He is unclear about where in New England this group resided at that time.

Maine

Trigger (1978:138) notes that the Eastern Abenaki in Maine utilized weirs at the time of European contact. Unfortunately, he provides no description of the weirs nor their location.

New Hampshire

Three stone weirs were described in the 1930's in the Winnepesaukee region; two "V" shaped and the third "W" shaped. The points were oriented downstream, and spaces were left at each point (Proctor 1930:41-42,49). Moorehead (1931:51) noted one year later that most of the weirs were removed for navigational purposes, rendering contemporary analysis impossible.

Massachusetts

A "Captain Gosnold" reported in 1602 that at Martha's Vineyard, "we found an old piece of a weare of the Indians to catch fish" (Rostlund 1952:170). A single stake found at the Buswell site on the lower Merrimack is claimed to be part of a prehistoric fishweir (Brumbach 1986:39; Barber 1980:97), but the claim is dubious, as the stake was left in situ untested, and no evidence for a prehistoric date is presented (Barber 1980:103).

The best known prehistoric weir on the eastern seaboard is the Boylston Street weir, located under the streets of Boston. About 65,000 stakes interwoven with brush wattling occur throughout an area of approximately two acres (Johnson et al 1942; Johnson 1949; Johnston and Cassavoy 1978:708). Recent investigations (yet unpublished) determined, contrary to previous reconstructions, that these stakes constituted numerous small weirs built over many centuries (Dena Dincauze and Elena Décima, personal communication). Like the stake weirs at Atherley Narrows, these features date to the Late Archaic [10], and seem to be tidal weirs (Dincauze 1973:31).

Connecticut

An unidentified number of stake weirs are known from the Housatonic River. Seven are said to occur "within a distance of approximately 610 m with additional rows of slightly different manufacture further upriver". They are described as consisting of "palisades of stakes about 5 cm (2 inches) apart across the mouth of the river" (Brumbach 1978:211)).

Rhode Island

Mourt's Relation, written by one of the colonists of the Plymouth Plantation, mentioned the use by Native Americans of a weir (type unspecified) near present-day Bristol (Heath 1963:63). There are no known accounts of extant weirs in Rhode Island.

New York (excluding Iroquois area)

A number of contact era accounts attest to the use of weirs in New York State. Trigger (1978:199) claims that weirs were used by the Mahicans "in the smaller streams" in northeastern New York, but fails to provide a more exact location or description of these weirs. Further downstream in the Hudson drainage, Native Americans employed an eel weir (no description is available) in Orange County in historic times (Beauchamp 1900:131). Weirs may have been utilized on the lower Hudson in the early seventeenth century when the Dutch first settled there. DeVries described the use of large seines, and also nets "set...on sticks in the river, one and one-half fathoms deep" (Brumbach 1986:42-43); the latter may refer to weirs, or simply weir-like devices.

A semi-circular stone weir is now or was once present on Fish Creek, at the outlet of Saratoga Lake (Rau 1885: 201). Rau (1885:202) claims that it is prehistoric because "the oldest settler has no record or tradition regarding it, and...there are directly upon the top of the wall, in different places, stumps of white oak betokening a growth of several centuries." Brumbach (1986:39) doubts that this is a prehistoric structure. If it is prehistoric, this would be the only weir of its type ever described from eastern North America.

In southern New York, the use of a brush weir by the Oneida to catch shad on the Susquehanna was described in the 1670 account of Dablon, a Jesuit missionary (Brumbach 1986:39). A late eighteenth century description from the same area noted: "...they tye Bushes together so as to reach over the River...(and) sink them with Stones" (Brumbach 1986:40).

Middle Atlantic (non-specific)

Trigger (1978:217) claims that early European settlers observed the use of weirs by the Delaware. He is unspecific about the exact region or type of weirs.

Pennsylvania

Rostlund (1952:170) mentions accounts of stone weirs in use on the Juniata and Lehigh Rivers (the latter in the vicinity of Bethlehem) written by European settlers in the early seventeenth century. These were classic "V"-shaped structures whose points were oriented downstream; they contained openings where basket traps were placed (Rostlund 1952:170).

In more recent times, stone weirs were sighted in the Schuykill River, but no descriptions or accurate locations are provided (Strandberg and Tomlinson 1969:319). Stone weirs may also exist on the Delaware in Pennsylvania, but the allusion to this possibility by Strandberg and Tomlinson (1969:319) is so unspecific that it is unclear to which part of this long river the authors refer. Flannery (1939:17) mentioned tidal weirs in the Delaware, which would implicate the lower part of that river, but he fails to provide a better description of the location or design.

New Jersey

In 1913, Skinner and Schrabisch (1913:75-78 listed eleven stone weirs in the Passaic River [11]; ten years later, nine of these still remained (Heusser 1923:23). All are in the shape of a "V" with the apex oriented downstream (Lenik 1985:26). One of these, the Fair Lawn/Paterson fishweir, remains visible to this day, and is under study by the author [12] (Figure 2). There is early historic evidence that these weirs are of prehistoric origin [13]. The weir spans the entire width of the river, and there is no opening at the apex as is described for many similar structures.

Kraft mentions two stone weirs in the Ramapo River (Bergen County Office of Cultural and Historic Affairs 1983:L11; Kraft 1976:7). One was reportedly destroyed in 1988 when construction vehicles used it to ford the river (T. Robin Brown, personal communication); the other is unlikely to be of prehistoric origin, based on personal inspections. This latter `weir' consists of a unique arrangement in the form of two "L"-shaped masses of rock (Kraft 1976:7; Marshall 1983:27); a large sawed wooden beam is deeply embedded within the structure.

Delaware

There is no clear evidence for prehistoric weirs in Delaware. Individual stakes were found in the last century in the Delaware River near Wilmington, but their association with a weir is tenuous at best (Holmes 1896; Connaway 1982:158).

Maryland

Thirty-six stone weirs were discovered in a survey of the Potomac River between Point of Rocks, Maryland and Harper's Ferry, West Virginia (Strandberg and Tomlinson 1969; Strandberg 1962) (Figure 3). A number of these are likely of colonial construction, but Strandberg and Tomlinson claim that some undoubtedly date to before European contact [14] (Strandberg and Tomlinson 1969).

Virginia

Ethnographic and contemporary accounts of weirs are perhaps more common for Virginia than for any other state. Trigger (1978:258,284) mentions that the Algonquians in this state employed weirs at the time of European contact, as did Iroqouian tribes on the Virginia/North Carolina border. He is unclear about the location or type of weirs utilized by the Algonquians, but does note that Iroquoian groups utilized "reed" weirs. Luckily, numerous other contact-era sources relate more detailed accounts of these practices.

Accounts of weir use in Virginia date back to the 1580's. M. Thomas Hariot and Ralph Lane observed the use of "reed" weirs at the time, but failed to describe such details as layout, location and method of employment (Hakluyt 1907:183). However, the engravings which accompanied Hariot's book (published in 1588) serve to ameliorate some of these shortcomings. The engravings, prepared by Theodore DeBry from John White's watercolors (painted between 1585 and 1586, and now lost) depict stake weirs on the coast near Roanoke. One of the weirs appears to be a simple reed fence (Figure 4), while another consists of interconnected enclosures that open one into the next (Figures 5 and 6) (Hariot 1588). One century later, these same weirs were still in use, as were other types, judging by the following account (Beverley 1705:148):

The larger Fish, that kept in deeper Water, they were put to a little more Difficulty to take; But for these they made Weyrs; that is, a Hedge of small riv'd Sticks, or Reeds, of the Thickness of a Man's Finger, these they wove together in a Row, with Straps of Green Oak, or other tough Wood, so close that the small Fish cou'd not pass through...(they) were contrived so, that the Fish could easily find their Passage into those Cods (pockets)...but not see their Way out again, when they were in; Thus if they offered to pass through, they were taken.
Sometimes they made a Hedge as this, quite a-cross a Creek at High-Water, and at Low wou'd go into the Run, so contracted into a narrow Compass, and take out what Fish they pleased.
At the Falls of the Rivers, where the Water is shallow, and the Current strong, the Indians use another kind of Weir, thus made: They make a Dam of loose Stone, where of there is plenty at hand, quite across the River, leaving One, Two, or more Spaces or Trunnels, for the Water to pass thro'; at the Mouth of which they set a Pot or Reeds, wove in Form of a Cone, whose Base is about Three Foot, and in perpendicular Ten, into which the Swiftness of the Current carries the Fish, and wedges them so fast, that they cannot possibly return.
In this description, Beverley describes three distinct types of weirs in use. The first consists of a stake trap into which fish simply wandered and were ensnared; this appears to be of the `labyrinthine' variety. The second is also a stake weir, but is a simpler tidal trap. The third is a stone weir on a river.

No stake ("reed") traps are known today from Virginia; indeed, there are no extant tidal traps anywhere along the Atlantic coast (Larson 1980:119). However, accounts of stone weirs in Virginia's rivers abound. Unfortunately, it is often difficult to distinguish prehistoric from historic weirs. Strandberg and Tomlinson (1969:319) described a number of stone weirs in this century in the James and Shenandoah Rivers. At least two of the James River weirs, in Richmond, are very likely of prehistoric origin, judging by their inclusion in a seventeenth century manuscript. They are "W"-shaped with the points oriented downstream (Bushnell 1930:8). In addition, there are mid-18th century references to a stone weir on the Otter River in Bedford County (northwest of Altavista) (Gravely n.d.).

Stone weirs are also known from the Rappahannock River (above Fredericksburg) and the Rapidan River (at "Skinkel's Ford") (Bushnell 1935). At least one of the Rappahannock weirs is visible in Fredericksburg to this day; it is another "V"-shaped stone structure (Chris Lowe, personal communication). There is also a stone weir ("V"-shaped) visible in the Smith River in Martinsville (Gravely 1973). There are reports of historic Native American weirs on the Rappahannock, but their design is unlike any other known prehistoric weirs, and they are confirmed to be of historic construction (Speck, Hassrick and Carpenter 1946).

North Carolina

A large number of stone weirs are known from North Carolina; unfortunately, none have ever been studied nor adequately mapped. Peck (1977:5) alleges that over one hundred weirs (certainly not all prehistoric) are said to be visible from the air in the Rocky River, but he fails to cite scholarly sources attesting to this fact. Better evidence exists for other, isolated weirs. William Oliver (personal communication) affirms the existence of a "V"-shaped stone weir at the Town Creek Indian Mound (31 Mg 3) in Montgomery County, another stone weir (no description provided) at the Trestle Site (31 An 19), and other stone weirs in the Pee Dee River in Montgomery, Richmond, Stanly and Anson counties. In a voyage down the Yadkin and Pee Dee Rivers in the 1920's, Douglas Rights (1929) encountered a dozen "V"- and "W"-shaped stone weirs. He claims (Rights 1929:18) that they are of prehistoric origin based on the proximity of "Indian village sites." Peck (1977) established that stone weirs on the Pee Dee River consist of four distinct types, including "V"-shaped with a simple opening, "V"-shaped with a "parallel" opening, "W"-shaped, and "Multiple V"-shaped. All of these types are described as having openings at the apexes.

Ethnographic accounts of weir use in North Carolina are rare. Contact-era sources mention that Algonquian tribes in this state used "reed" weirs (Trigger 1978:273). In his diary, Ralph Lane discusses the use of weirs (of an unspecified type) on the coast in March of 1586 (Larson 1980:121).

South Carolina

There is a single ethnographic report of a weir in use by Native Americans from 1765 on the Little River (Merriweather 1940:169). It is unclear what type of weir this was, and whether or not it was in use previous to European colonization.

Four stone weirs exist in the Trotters Shoals Reservoir area. Investigations in the early 1970's focused on three of these structures [15] (Hemmings 1970). The first, 38 Ab 8, is located on the Savannah River at its juncture with the Calhoun Branch. It consists of a 400 foot-long wall of boulders paralleling the stream bank, with another 200 foot-long wall angling away from it as one follows it upstream. The only evidence that it may be prehistoric is the close proximity of a prehistoric site (38 Ab 10) (Hemmings 1970: 49-50). The second, 38 Ab 16, consists of an irregular boulder alignment, approximately 200 feet long, with openings at irregular intervals. This structure is also in the vicinity of a known prehistoric site (Hemmings 1970:50). The third, 38 Ab 15, is near Carter Island on the upper Savannah River, and it too is associated with a prehistoric site in the immediate vicinity. A 300 foot-long irregular wall of boulders contains two "V"-shaped alignments near the center of the feature. Logs incorporated into the feature were tentatively radiocarbon dated to ca. A.D. 1400 [16], with a rebuilding episode around 1770 (presumably by European colonists) (Hemmings 1970:50; Hemmings 1973:44).

Alabama

It is clear that stone weirs in the Tallapoosa River are of prehistoric affiliation (Hubbert and Wright 1987). There, Hubbert and Wright (1987:8,51) discovered over 50 stone weirs, mostly located on shoals. Four types were distinguished: Single "V", Double "V", Triple "V", and straight (Hubbert and Wright 1978:59-61) (Figures 7 and 8). They were deemed to be of prehistoric origin, based on criteria which included: 1) a lack of historic settlement due to inaccessibility (except by canoe) (Hubbert and Wright 1978:ix); 2) a large number of prehistoric sites, whose locations were very strongly correlated with the locations of weirs (Hubbert and Wright 1978:96); and 3) a lack of exploitable resources aside from fish (Hubbert and Wright 1978:95).

Two interesting types of feature accompanied these weirs. The first was the presence of beds of burned rocks at sites associated with weirs. These features were interpreted as the locations of drying racks for the preservation of fish (Hubbert and Wright 1978:99) [17]. The second type of feature consisted of piles of rocks of the type utilized in the construction of the weirs. Hubbert and Wright (1978:41) claim that these features "could represent the raw materials from which the weirs were constructed." Based on analysis of the nearby sites, these weirs are said to date back to the Middle Archaic, with the associated sites being occupied for only brief periods (Hubbert and Wright 1978:99-100).

Tennessee

In the only early account describing Native American weir use in Tennessee, Lieutenant Henry Timberlake, in his memoirs (1756-1765), recounted the use of a "V"-shaped weir. He did not describe the materials used (Williams 1927:69).

An extensive survey in 1985 of large stretches of the Tellico and Little Tennessee Rivers revealed that no weirs were extant (Chapman 1985). Myer (1928:26,782) and Swanton (1946:335) identified "V"-shaped stone weirs in the Tennessee River in Knox County, and in the Obey River (near the mouth of Eagle Creek) in Pickett County. Stone weirs in the lower Holston River were firmly dated to the nineteenth century (Cobb 1978).

Georgia

Chapman documented 37 stone weirs were in a 45 mile stretch of the Etowah River between Cartersville and Rome, but admitted that a number are historic (Chapman 1975:6). They consist of four basic shapes: curved, "V"-shaped (with one or more openings), "L" shaped (with openings), and irregularly shaped (Chapman 1975:8). Chapman (1975:6-7) describes a dichotomy between weirs utilizing large rocks and those with small. Embedded historic materials appear to be associated solely with the former.

George Henry Loskiel described a "V"-shaped stone weir in use by Native Americans in 1794: "(T)he Indians run a dam of stones across the stream, where its depth will admit of it, not in a straight line, but in two parts verging towards each other in an angle. An opening is left in the middle for the water to run off. At this opening they place a large box, the bottom of which is full of holes" (Jones 1873:332-3).

Kentucky

The only reference regarding weirs in Kentucky is mention of a stone weir in Bourbon County. A more exact location and description are unavailable (Funkhouser and Webb 1932).

Mississippi

Only a single weir is known from this state, the Sturdivant Fishweir, in the Homochitto River in Amite County. This feature was the site of extensive archaeological investigation, and is the best preserved weir of its type (Connaway 1982). It consisted of rows of stakes (314 were recovered, most of Yellow Pine), clearly worked with stone tools, which were interwoven with split cane mats. The weir is in the shape of a "V", with a gap at the apex. Parallel rows of stakes attest to episodes of rebuilding, and the stakes were radiocarbon dated to the Late Mississippian/early Protohistoric period (Connaway 1982:138-152) [18].

Florida

A number of sixteenth-century accounts describe the use of tidal weirs on the Atlantic and Gulf coasts of Florida. The earliest such account is probably that of Garcilaso. He described what apparently was a stone tidal weir at Tampa Bay utilized to catch rays (Swanton 1946:333). Jean Ribault and Sir John Hawkins described the use of stake weirs in the 1560's (Larson 1980:121; Connaway 1982:159). Ribault noted that the weirs were "built in the water with great reeds, so well and cunningly set together after the fashion of a labyrinth, with many turns and crooks" (Connaway 1982:159). Laudonnière described these same weirs as "inclosures, made of reeds, and framed in the fashion of a labyrinth, or maze..." (Laudonnière 1869:180). They are depicted in Le Moyne's paintings, which date to the 1560's (Lorant 1946:41) (Figures 9 and 10), but none of these weirs are reported to be extant.

Analysis of weir use

Geographic variability

There appears to be little, if any, correlation between weir design and geographic province. Tidal weirs made of stakes are known from Canada to Florida, and other stake weirs are known from Maine to Mississippi. Stone weirs in riverine settings, most often of the "V"-shaped variety, are known from nearly every state from New York to Georgia. Tidal weirs of stone are reported from Florida only, but there is an ambiguous reference to tidal weirs in the Delaware River (in Pennsylvania) which may also be of stone.

Methods of use

The use of tidal weirs is relatively straightforward. There is general agreement among all accounts that a barrier is positioned so as to trap fish behind the falling tide (Rostlund 1952:101). Unfortunately, there are no detailed descriptions of tidal weirs. However, early engravings and paintings depict either straight walls perpendicular to the shoreline, or "labyrinthine" structures. Based on these depictions, tidal weirs seem to have operated in a passive fashion.

Descriptions of the use of riverine weirs vary remarkably, and are often times contradictory. Most agree that these weirs are generally associated with the exploi-tation of anadromous and catadromous fish (Schalk 1977:232).

Discussions of riverine weirs place great emphasis on their operation during spawning runs, generally in the spring and summer along the Atlantic (see, for example, Bakeless 1961:223,262; Brumbach 1986:42; Beverley 1705:146). Many stone weirs throughout eastern North America are described as having openings for traps, invariably at the downstream-pointing apexes (for example, Connaway 1982:156; Rostlund 1952:170; Myer 1928:782; Chapman 1975:8). This is odd, as fish traveling upstream would be diverted by the stone walls to the sides of the river, not to the openings in the center.

In fact, there are accounts of the use of weirs to capture fish as they traveled downstream (aside from eels, which migrate downstream to spawn) in both active and passive manners. Beverley, in a quote related previously, described the passive operation of such a weir, whereby "the Current carries the Fish, and wedges them so fast, that they cannot possibly return" (Beverley 1705:148). The operation of weirs in this fashion is known historically from southeast Alaska (Moss and Erlandson 1990:144). More common are descriptions whereby fish are driven into the apex of "V"-shaped weirs through agitation of the water upstream (Williams 1927:69; Chapman 1975:3).

Historic weirs served to capture fish traveling both upstream and downstream (Reed 1979), and it is clear that native Americans made use of this method in some places (Wheeler and Jones 1989:168). For example, the Jesuit missionary Dablon described the capture of both anadromous and catadromous species with the same weirs in New York (Brumbach 1986:40). Similarly, the Micmac would "leave an opening in which they place a bag net so arranged that fish run into it. In autumn, when the fish descend, the opening of the bag is in the other direction" (Rostlund 1952:170). Johnston and Cassavoy (1978:704) allege that the Atherley Narrows weirs appear to be oriented to capture fish migrating upstream, and indeed, fish remains are found at local sites; however, previously cited contact-era accounts attest that the most important species caught there was eel, which migrate downstream to spawn.

It is evident that, despite a focus on the use of weirs to capture anadromous fish passively on their upstream migrations, riverine weirs were employed to capture fish traveling downstream as well, and furthermore were employed in an active fashion. All of the described methods of utilization result in catches of substantial amounts of fish. Since weirs were used throughout the entire eastern seaboard of North America, the widespread exploitation of fish certainly carries implications for subsistence patterns throughout the region. Certain other practices related to subsistence are implied as well.

Implications of weir use

Fish processing

Weirs are designed to capture fish in large numbers. Since fish spoil very rapidly, a knowledge of preservation is an essential accompaniament to weir utilization. The extensive use of weirs in eastern North America indicates that a knowledge of preservation was widespread throughout this region (Schalk 1977:232).

Wheeler and Jones (1989:28) characterize the preservation of fish as a "relatively easy" task, involving drying, salting, smoking, fermenting, or a combination of these. However, the exact application of these methods was diverse, and may have been species and/or size specific (Whyte 1988:115).

A great diversity of specific methods was employed across the varying regions and among different groups in eastern North America. Rostlund (1952:195-6) observes that the methods known to be in use at the time of European contact included drying, presumably by fire (among the Delaware, Chitimacha, and Atapaka, and in the St. Lawrence River and Apalachee Bay areas), sun-drying (along the lower Hudson River), fire-drying (in Virginia), a combination of sun-drying and smoking (among the Hurons and Narraganset), smoking alone (among the Natchez, and among other groups in the Roanoke Islands and South Carolina mainland), drying, smoking, and freezing (among the Ojibwa), and a combination of drying and salting (among the Yuchi).

There are isolated examples of more unusual methods of preservation. The Hurons, Mohawk, and other Iroquoian tribes ate powdered dried fish (Rostlund 1952:199). The Hurons also prepared "sagamite," or spoiled fish, apparently preserved through natural fermentation (Brumbach 1978:254). "Fish preserves," whereby live fish were impounded in artificial canals and ponds, are known from DeSoto's expeditions (Jones 1873:325-6). And one site near the Delaware River in New Jersey (in the vicinity of Trenton) is interpreted as the location of a vast complex for rendering fish oil (Louis Berger and Associates, Inc. 1987:VIII-5) [19].

Length of occupation of weir sites

The availability of large quantities of fish in a restricted area increases the opportunity for sedentism (Schalk 1977:232). In addition, the chores (setup and maintenance of facilities, processing of catch, etc.) necessitated by weir use require a certain level of (at least seasonal) sedentism (Avery 1975:113; Brumbach 1986:35,37; Larson 1980:125; Schalk 1977:232). We do not yet have the information to describe typical weir use, but it is still possible to reconstruct the potential minimal time during which a weir site may have been occupied. The most important variable for this reconstruction is the availability of fish (Brumbach 1986:35).

Seasonal availability of fish

Anadromous spawning runs
A number of authors who concentrate on fish as a subsistence item place great emphasis on the seasonal spawning habits of anadromous fish. This is no doubt due to the convergence of considerable numbers of large fish in a short time-span, which makes this food resource easily obtainable in a small area and over a short period of time (Fisher 1983:36; Schalk 1977:213; Wheeler and Jones 1989:5).

In preceding discussion, I noted that weirs were not employed exclusively or predominantly for capturing anadromous species during their upstream migrations. However, anadromous fish are a qualitatively superior food resource during their spawning runs, at which time their fat reserves are highest (Brumbach 1986:36). The high fat content of anadromous species rendered them attractive, and sometimes necessary, to a diet which was likely low in fat during the winter in some regions, particularly the Northeast (Speth and Spielmann 1983:2).

The specific mechanisms regulating anadromous spawning are not yet completely understood, particularly regarding the environmental conditions which determine the timing of their runs. There are two predominant schools of thought regarding the mechanisms governing spawning runs: The first points to water temperature as the overriding factor initiating spawning runs (Schalk 1977). An illustrative example is the spawning habits of the shad (Alosa spp.), which adhere to similar temperature regimes by following the isotherms or thermal zones in the range of 13º to 18º C. By following these zones the fish arrive at the rivers at the proper time for migration and spawning: In their southern range the shad enter the St. Johns River in Florida in January when water temperature has cooled to around 15ºC. Further north, the spawning runs occur progressively later in the year when the waters have warmed adequately [20]. (Brumbach 1978:224) Water temperature is not necessarily determined solely by drops in ambient temperature due to the season. Observations from California demonstrate that spawning runs may be initiated by water cooled both in the autumn (through ambient temperature change) and spring (due to melting snow from nearby mountains) (Swezey and Heizer 1984:970). This appears to be the case only in larger rivers (Swezey and Heizer 1984:971-2).

The alternate explanation of anadromous migrations is that they are regulated by the volumetric flow of water from the rivers in which anadromous fish spawn (Banks 1969). Although the exact reasons for this are not clear, Banks claims that "many factors like water and air temperature, turbidity, atmospheric pressure, cloud cover, pH and variations in concentrations of many dissolved ions are associated with the rate of water discharge" (Banks 1969:86). This phenomenon was confirmed through a number of controlled experiments which were successful regardless of water temperature, leading Banks (1969:120) to conclude that "(t)he evidence of the influence of temperature on upstream migration is both conflicting and inconclusive." He notes, though, that this phenomenon may apply only to smaller streams, and that "(e)arly in the season, temperature may be the factor controlling migration" (Banks 1969:107,115).

Regardless of the factor initiating spawning runs, it is clear that different species spawn at different times of the year; therefore, the higher the diversity of species present, the greater the time period during which fish can be exploited in abundance (Schalk 1977:216,241). A number of factors influence species diversity; diversity is more restricted in northerly latitudes, and is greatest south of 45°. The size and section of a river also strongly influence species diversity. Larger rivers support greater numbers of species, their lower sections being the most productive.

Surrounding topography as well plays a part in diversity. Rivers in estuary or upwelling zones are most suitable, especially when they are associated with wide continental shelves at the coast (Schalk 1977:218,241; Perlman 1980:262-3).

Observed availability
It would do well at this point to look at the actual observed availability and exploitation of fish on a regional basis to add to the theoretical knowledge explicated above. As previously pointed out, the contemporary situation has been compromised too greatly for direct comparison with the prehistoric period. Research derived from contemporary studies is cited with the understanding that prehistoric availability either met or exceeded modern yields, and when possible, contact-era and other historic accounts are cited for reconstructing past availability. Each of the species discussed below is known (through either contact-era accounts or archaeological investigation) to have been exploited by various native American groups.

Judging by numerous contact accounts, fish were available throughout most of the year from the Great Lakes/Finger Lakes to the Canadian Maritime provinces. These accounts tend to be the very early accounts of Jesuit missionary and pioneer explorers in the region. Pierre Biard, a Jesuit missionary near present-day Annapolis, Nova Scotia, wrote in 1616:

In the middle of March, fish begin to spawn, and to come up from the sea into certain streams, often so abundantly that everything swarms with them. Any one who has not seen it could scarcely believe it, you cannot put your hand into the water, without encountering them. (Brumbach 1986:42)
and
From the month of May up to the middle of September, they are free from all anxiety about their food; for the cod are upon the coast, and all kinds of fish and shellfish...In the middle of September (they) withdraw from the sea, beyond the reach of the tide, to the little rivers, where the eels spawn, of which they lay in a supply; they are good and fat. (Snow 1980:43)
Fish continued to be available through the fall (Johnston and Cassavoy 1978:707), during which time Samuel Champlain observed a group of Iroquois fishing (in October) (Brumbach 1978:219). Because of the diversity of species, fish are available year-round in this region, and at least one group, the Huron, are known to have exploited this resource throughout the entire year (Heidenreich 1971:208).

We have fewer early descriptions from New England from which to reconstruct subsistence patterns. We know, based on descriptions provided by the Plymouth colonists, that Native Americans fished for alewife in April and eels in September in Massachusetts (Bakeless 1961:22-3). Sturgeon were also plentiful in that state, and were exploited during their spawning runs in the Spring (Barber 1980:107). Other species which still make spring spawning runs in New England to this day include the shad and salmon (Brumbach 1986: 47,60). Of course, freshwater species are also available; the Plymouth colonists noted that bass fishing was practiced in the summer in the vicinity of Rhode Island (Heath 1963:63).

Contact-era sources from the Middle Atlantic region which describe native fishing with any detail are almost nonexistent. Based on contemporary studies of known native species and a few historic descriptions, however, we may infer that fish were widely available throughout the year. Van der Donck, describing Native Americans fishing on the lower Hudson, alluded to this in an inexact fashion when he wrote:

First in the fishing season they caught many shad... Later they caught striped bass...Later still they caught the drums...For those fishes succeeded each other in their seasons...There are also carp, snook, forrels, pike, trout, suckers, thickheads, eels, palings, brickens, and lampreys. (Van der Donck 1656:54)
A number of anadromous species were available in this region, including sturgeon, herring, alewives, shad, and eel, most of which made significant spawning runs beginning in the spring and were available through the summer (Brumbach 1978:229-234; Brumbach 1986:50-1). In addition, striped bass were available throughout the year in the lower Hudson River and its tributaries, as were numerous freshwater (non-anadromous) species (Brumbach 1978:20,233). Similarly, two species of sturgeon (Acipenser brevirostrum and A. oxyrhyncus, also known as A. sturio) were available in the Delaware and Hudson Rivers year-round (Brumbach 1978:226).

Detailed contact-era and early historic accounts of native fishing practices are similarly lacking for southeastern North America, but as with the Middle Atlantic region, contemporary studies of non-introduced species can aid us in reconstructing the prehistoric availability of fish. Previously mentioned anadromous species (especially the sturgeon, herring, and shad) were all procurable beginning in the spring, and were certainly exploited (Brumbach 1978:227; Larson 1980:112-3; Brumbach 1986:47; Hakluyt 1907:182; Jones 1873:324). Because of the topography and species diversity, anadromous fish were accessible for an extended period of time (Schalk 1977:216,241; Perlman 1980:262-3). In addition, non-anadromous species, both freshwater and ocean species, were available and exploited throughout the rest of the year (Chapman 1975:2; Larson 1980:81-95,123; Hakluyt 1907:182; Whyte 1988:115). Early historic accounts attest to winter fishing in Florida (Bennett 1975:15).

The role of fish in reconstructing prehistoric subsistence and settlement patterns

Northeast

In the Northeast, hunting, gathering and fishing constituted the primary subsistence activities throughout nearly all of prehistory. This basic pattern persisted until the Late Woodland period, at which time plant domestication became widespread (Brumbach 1978:163). Excluding the `Maritime Archaic Tradition' on the east coast of Labrador, the earliest known fishing on a large scale in the Northeast was during the `Brewerton Phase' in the Late Archaic in central and northern New York Sate. However, there is little archaeological evidence for the Early and Middle Archaic periods (Funk 1983:317,323). The Late Archaic period is generally characterized by the existence of hunting/fishing stations in the spring and summer, with settlements relocating further inland, where hunting predominated, during the fall and winter (Funk 1983; Kraft and Mounier 1982:61; Brumbach 1978:164). Overall, evidence for fishing is slight for both this period and the succeeding Transitional Phase (Funk 1983:331-332).

Sites throughout the Archaic are usually small, with larger sites being interpreted as evidence of repeated seasonal use (Ritchie 1956:74). The Early and Middle Woodland are similarly characterized by small, seasonal camps, situated near waterways in the warmer months and further inland during the fall and winter (Funk 1983:337,344; Williams and Thomas 1982:122).

Overall, the archaeological literature fails to emphasize fishing as an important resource. However, the data from individual site reports, as well as information regarding general settlement trends, hint at the importance of fish in the diet of prehistoric North Americans. Interestingly, most major sites appear to be located adjacent to waterways where considerable quantities of fish are attainable (Brumbach 1978:164; Bradley 1987:15; Trigger 1978:324), and many are situated exactly at those points most conducive to fishing (Brumbach 1986:37). Sites along the Charles River in Massachusetts date progressively earlier as one moves downstream. Dincauze (1973:25) interprets this pattern as evidence that ideal fishing locations were the basis for situating sites, and that sites needed to be relocated upstream as post-glacial sea transgression caused suitable fishing stations to be located further inland. The Dougall Site, near the Atherley Narrows weirs, contains extensive evidence of being a provenance for fish exploitation. Cultural deposits at that site date from the Middle Woodland through the Late Woodland (Johnston and Cassavoy 1978:707). As was previously pointed out, the weirs themselves date back to the Late Archaic, as do those at Boylston Street. In fact, a few sites demonstrate evidence of large-scale fishing dating as far back as the Early Archaic and even Paleo-Indian periods (Gramly and Funk 1990:24) [21].

Southeast

Evidence for the role of fishing, as well as overall reconstructions of prehistoric subsistence and settlement patterns for the Southeast, are meager. This is especially true for the pre-Woodland periods. Most authors of regional overviews see subsistence and settlement hinging on seasonal migrations between areas offering differing resources throughout prehistory. When fishing is mentioned, it is seen as a short-term occupation (generally during the summer) (Jones 1873:324-325; Jenkins and Krause 1986:41; Hubbert and Wright 1987:99).

Sedentism in the Southeast preceded sedentism in the northeast, appearing by the Late Woodland period "in the interior River Valleys in (the) mid-South" (Anderson and Garrow 1988:133-5). Most authors ascribe sedentism to the increased exploitation of plant and animal resources, primarily deer and nuts. Aquatic resources are occasionally mentioned as an element correlating with increased sedentism (Anderson and Garrow 1988:136,155). Still, a sedentary lifestyle did not become widespread until much later. In discussing the reasons for a lack of sedentism in this region, Jenkins and Krause (1986:119) point out:

A continuing need for fresh meat and skins for blankets, robes, and clothing could not be...satisfied. Nor could the local reserves of collectible non-nut foods be efficiently utilized...The local game animals and collectible non-nut foods would soon be exhausted by a fully settled population...
Jenkins and Krause omit discussion of the many factors which could alleviate many of these shortages, such as domestication of plants, the ritual or purposive regulation of resources
[22], and a higher availability of resources (especially fish) than is generally assumed.

The importance of fish as a resource is well documented at some southeastern sites (e.g., Jenkins and Krause 1986:69), as is evidence of the exploitation of a great diversity of species (e.g., Whyte 1988:115). Seasonal availability of fish in the Southeast is the least constrained of all areas on the Atlantic seaboard, providing not only the most diverse anadromous species, but also numerous freshwater and saltwater species, for which evidence of prehistoric utilization is presented in section 6.2.1.2 (beginning on p. 51).

In light of this evidence, and the fact that far more weirs are known from the Southeast than from the Northeast, it is clear that more research is needed before concluding that short-term seasonal occupation characterized the predominant modes of settlement and subsistence in this region [23].

Conclusions

The use of weirs remains poorly understood despite the existence of numerous ethnographic descriptions of their use and the remains of a large number of extant weirs. In addition, the implications of their use have yet to be fully incorporated into general reconstructions of subsistence and settlement patterns for the eastern seaboard of North America.

The oversight of the importance of weirs stems in part from a lack of preservation in the archaeological record. The limitations imposed by contemporary recovery techniques further exacerbate this problem. A shortage of available historic information further compound this difficulty. Underestimations of the importance of fish as a food resource contribute to our inadequate understandings.

Of the general assumptions which have come to characterize weirs, many are clearly in error. Foremost among these is the assumption that weirs are only useful for short periods of time, i.e., seasonally. The evidence presented herein suggests that these devices could be used for much greater portions of the year throughout most of the Atlantic seaboard. Ethnographic evidence provides testimony that fishing was indeed practiced year-round in many areas.

Another characterization of weirs is that their primary importance is for passively obtaining anadromous fish during their upstream spawning runs. Numerous ethnographic descriptions suggest that they were quite often used in an active fashion, and descriptions of their layout suggest that fish were more often caught in stone weirs during downstream movements, rather than upstream. This applies not only to the capture of catadromous species (eels) during downstream spawning runs, but also to anadromous species in their post-spawning phase, and to non-anadromous species either swimming with the current or forced downstream by agitation of upstream waters.

Flannery's characterization of weirs as `undiagnostic' (1939:17) appears to be correct. Weirs of two basic types (stake and stone) occur throughout the entire Atlantic region, and there are no observable regional differences in design. Stone weirs are usually "V" or "W"-shaped, with apexes always aligned downstream, and they occur almost exclusively in riverine settings. Stake weirs occur in both riverine and estuarine/coastal areas, but too few survive from which to draw firm conclusions regarding similarities and differences in design. Without a way of dating stake weirs, it is difficult as well to determine whether there are specific temporal design characteristics.

The decision to use weirs affected settlement and subsistence strategies. Weir use (and the accompanying responsibilities ancillary to their use, such as weir set-up/maintenance and fish processing) indicates at least semi-sedentary settlement. Since large catches are easily obtained through weir use, and weir use was widespread, it follows that fish were exploited on a large scale across much of eastern North America at least as far back as the Late Archaic. The lack of archaeological evidence attesting to aquatic resource procurement does not necessarily reflect a concomitant deficiency in the availability or exploitation of this particular resource.

Because weirs are known from throughout the Atlantic coast, there is potentially a need for an overall re-evaluation of the general subsistence and settlement patterns, throughout the entire prehistoric period, for this immense area. This evaluation can only be accomplished if more light is shed on the nature of the construction and use of weirs and their temporal affiliation. In addition, a more thorough understanding of aquatic resource procurement and exploitation is required. The latter is a problem which will likely be better understood as archaeological recovery techniques improve. The primary hindrance in this realm stems from a lack of expertise in recovering and analyzing the osteological remains of aquatic species, a situation which seems to be improving (Wheeler and Jones 1989).

In addition to concentrating on aquatic resources in general, archaeological sites which appear to be linked to weir use in particular need to be examined, with emphasis placed on the retrieval and analysis of remains at a very small scale (to enhance the recovery of fish bones, scales, etc.).

Weirs themselves are seldom the focus of archaeological investigation. Few weirs have been sketched, and fewer still have been accurately mapped. Since weirs are more highly subjected to natural and human destruction than subsurface archaeological sites, the opportunity to examine them is quickly vanishing. Hopefully, a comprehensive program to accurately locate, sketch, and map these features will commence in the near future. Granted, there are logistical difficulties involved, considering that almost all of these structures are either partially or completely submerged; however, present technology (including aerial survey and the use of sonar for underwater mapping) should be sufficient to accomplish the task.

Notes

1. The following are excerpts of contact-era accounts attesting to this fact.

From New England:

Yea, when a heape of stones is reared up against [the alewives during their spawning runs] a foot high above the water, they leape and tumble over and will not be beaten back with cudgels. (Bakeless 1961:223; attributed to "Puritan settlers.")
[T]he greate smelts passe up [the Smelt River, near Plymouth, Massachusetts] to spawne likewise in troupes innumerable, which with a scoupe, or a boule, or a peece of barke, a man may cast upon the bank... (ibid.; attributed to colonist John Pory.)
[D]uring one month the fish ascend the river in so great numbers that a man could fill fifty thousand barrels with them in a day, if he could be equal to that work. (Brumbach 1986:42; from the 1723 account of Râle, a Jesuit missionary to the Abenaki.)
From the Middle Atlantic:
There are more records of shad [in the Susquehanna River] than of anything else, though there must have been just as many fish of other kinds. Two thousand shad were taken in a single night at Wyalusing, Pennsylvania, but that was nothing compared with the celebrated `widow's haul' on the North Branch, which brought in ten thousand at once... (Bakeless 1961:262; describing the travels {in Pennsylvania} of Conrad Weiser, a Moravian missionary in the early 18th century.)
The Indians come up this river in canoes to fish, because it is one of the richest fisheries they have. (De Yoe 1961; quoting Jasper Danckaerts, a Labadist missionary in the vicinity of the lower Hudson River in March, 1680.)
From the Southeast:
In the Spring of the Year, Herrings come up in such abundance into their Brooks and Foards, to spawn, that it is almost impossible to ride through, without treading on them...Thence it is, at this Time of the Year, the Freshes of the Rivers, like that of the Broadruck, stink of Fish. (Beverley 1705:146; based on late seventeenth century observations.)
For quotes attesting to the availability and exploitation of fish on a grand scale in Canada, see the quotes in the section entitled, Observed availability.

2. Anadromous fish are those which migrate upstream from the sea to spawn; conversely, catadromous fish (primarily eels) are those which swim from lakes, rivers and streams out to sea to spawn (McClane 1978).

3. See, for example, Godwin 1988 on weirs in Australia; Avery 1975 for weirs in South Africa; Moss & Erlandson 1990 and Langdon, Reger & Wooley 1986 for weirs in Alaska. Numerous studies detailing the importance of fishing in the Pacific Northwest make mention of fishweirs; e.g., Moss & Erlandson 1990; Swezey & Heizer 1984.

4. These are the Sturdivant weir in Mississippi [Connaway 1982], various weirs in the Tallapoosa River in Alabama [Hubbert & Wright 1987], the Atherley Narrows weirs near Ontario, Canada (Johnston & Cassavoy 1978), and the Boylston Street weirs in Boston, Massachusetts (Johnson 1949; Johnson et al. 1942). The latter two are tidal traps.

5. For an example that fishing is not considered to be an aspect of hunting, see Lee & Devore's classic Man the Hunter (1968), which is typical in its emphasis on hunting land-animals, to the exclusion of aquatic resources.

6. It is not necessary that the entire device be permanent, but rather that significant portions remain in place. For example, it was noted in Alaska that the native Salish employed framework weirs from year to year which utilized removable latticework sections (Moss & Erlandson 1990:144). Ethnographic descriptions presented later in this thesis often make mention of removable traps, such as boxes, from stone weirs in riverine settings.

7. Larson (1980:123) claims that, for this reason, tidal weirs in the Southeastern U.S. only existed at Cape Hatteras, North Carolina and in Florida, but he is clearly unaware of the description (to be related later) by Beverley of a tidal trap in Virginia. In addition, it is not at all clear that the coastal weirs in use in North Carolina were tidal traps, as they were built perpendicular to the shore (ibid. p.120; discussion of which is also included later in this thesis).

8. For example, weirs are reported from southeast Alaska which consist of rows of sharp stakes, upon which salmon impale themselves when attempting to hurdle these obstructions during spawning runs (Moss & Erlandson 1990:143-4).

9. These are the (presumably uncorrected) radiocarbon dates listed in Table 1 on page 704:


10. The following (presumably uncorrected) dates are reported by Dincauze (1973:29):


11. They are as follows (Skinner & Schrabisch 1913):

page  NJ State Museum #   Description
 77      28 Be 82            Just below Wagaraw bridge
         28 Be 176/28 Pa 147 Fair Lawn/Paterson weir
 75      28 Pa 104           Fairfield, "a little below the county bridge"
 76      28 Pa 117           Little Falls, above High Bridge
         28 Pa 94              "      "      "     "    "
         28 Pa 101             "      "  below High Bridge running
                              from its south bank to Laurel Grove Island
         28 Pa 15 (?)        Paterson, below Main St. bridge
         28 Pa 16            Paterson north of Bunker Hill, along the
                              northernmost course of the river
 77      28 Pa 22            Paterson, between Broadway & Wessel bridges
 78      28 Pa 94            "Dundee Dam"
         28 Pa 96            "Belmont", from east bank to island
                             Opposite Belmont

12. This is the only stone weir confirmed to be extant in northeastern North America. It is located approximately 300 yards north of the Fair Lawn Avenue/Fifth Avenue bridge between the City of Paterson and the Borough of Fair Lawn.

13. Although direct proof of their affiliation with prehistoric peoples does not exist, overwhelming evidence suggests that these weirs predate European colonization in the area. The area was called "Sloterdam" by the earliest (Dutch) settlers there. The word means "Sluice Dam," and describes the resemblance of the weirs to vaguely similar-looking structures in Holland (Rogers 1960; Bergen County Office of Cultural and Historic Affairs 1983). The earliest known written occurrence of the term is in 1708 (Rogers 1960:12), which indicates that the term was probably in use prior to the beginning of the eighteenth century, at which time European settlement in the area was extremely sparse. By the latter part of the century, use of the term was in decline. A 1764 Road Return notes that the "Road of Slotterdam," a thoroughfare traversing the length of the Passaic River on its eastern side, was by that time "now known as River Road" (Bergen County Road Returns, Folder B9). The appearance of the term on Erskine's maps (1778-9) is one of the latest known uses (Smullen 1921).

Another reason to believe that the weirs predate European settlement is the total lack of historical reference to them. The area was thinly populated by Dutch farmers through the middle nineteenth century, and there is no mention in historic records of fish as a resource in this area during the colonial period. The only known historic use of fish weirs in the entire state before 1800 was by John Read, an agriculturalist from Burlington (far to the south). In 1763 he began a commercial fishing operation on the Delaware River, employing weirs constructed of mats and nets which in no way resembled the stone weirs found in the Passaic (Woodward 1941:399-400).

14. Part of the evidence for this claim derives from the fact that some of the weirs are "deep underwater" (Strandberg & Tomlinson 1969:312), and it is reported that "one of them was in use in 1724, when a Mr. William Nelson, the first white settler, visited the area" (Strandberg 1962:478).

15. The fourth, known as Fishdam Ford, is located on the Broad River near the town of Carlisle. No known description of this features exists.

16. Only two dates were procured; the first was 545 ± 100 B.P., the second 180 ± 80 B.P. (Hemmings 1973:44).

17. Interestingly, Charles Cobb (personal communication) has observed similar features in association with natural shoals in the Georgia.

18. The dates are as follows [Connaway 1982:152]:

19. The unique interpretation of this site (the Trenton Complex, Area B {28 Me 1-b}) is based on 1) the presence of anadromous fish (indirectly inferred from high levels of strontium and mercury in the soil); 2) a lack of other exploitable resources in the vicinity; 3) the presence of boiling-stone heaps as a predominant feature; and 4) analogy with Northwest Coast practices (Louis Berger & Associates, Inc. 1987).

20. This raises an interesting point: if Florida's waters need to cool, and northerly waters need to warm up, this seems to imply that there is there a point in between where the proper temperature for spawning exists year-round.

21. Examples include the Kipp Island site in central New York, where freshwater fish made up the bulk of faunal remains (Trigger 1978:324); the Buswell site in the lower Merrimack Valley in Massachusetts, where a nearly continuous 7,000 year occupation (from Middle Archaic through Late Woodland) is interpreted as "a settlement used for the exploitation of spawning anadromous fish," based on its unique topography, and faunal and artifact analyses (Barber 1980:97); and the Neville site, at Amoskeag, New Hampshire, where fish were being exploited nearly continuously from 8,000 B.P. onward (Dincauze 1976).

22. See Swezey & Heizer 1984 pp. 981 and 985 for examples of the regulation of anadromous resources in Northern California.

23. At least one example of a semi-sedentary site based on the exploitation of fish is known from contact-era accounts: Early Spanish chroniclers, including Garcilaso de la Vega, noted that in the town of Hirihigua (also known as Ucita), a semi-sedentary site in or near Tampa Bay, Florida, little or no maize was available, but fish were an important resource, and weirs were heavily utilized (Larson 1985:125).

Sources Cited

Anderson, David G. and J.W. Joseph Garrow & Associates, Inc., 1988:Prehistory and History Along the Upper Savannah River: Technical Synthesis of Cultural Resource Investigations, Richard B. Russell Multiple Resource Area, Vol. I. Garrow & Associates, Inc., Atlanta.

Avery, G., 1975: Discussion and use of tidal fish-traps. South African Arch. Bull. 30:105-113.

Bakeless, John, 1961: America As Seen by its First Explorers: The Eyes of Discovery. Dover Pubs., New York.

Banks, J.W., 1969: A review of the literature on the upstream migration of adult salmonoids. J. Fish Biol. 1:85-136.

Barber, Russell J., 1980: Post-pleistocene anadromous fish exploitation at the Buswell site, northeastern Massachusetts. In Early and Middle Archaic Cultures in the Northeast (D.R. Starbuck and C.E. Bolian, eds.). Occ. Pubs. in Northeastern Anthro. No. 7, Franklin Pierce College, Rindge, NH, pp. 97-113.

Beauchamp, William M., 1900: Aboriginal Occupation of New York. Bulletin of the New York State Museum 32(7), University of the State New York, Albany.

Beauchamp, William M., 1905: Aboriginal Use of Wood in New York. N.Y. State Mus. Bull. #89, pp. 87-272. New York State Education Dept., Albany.

Bennett, Charles E., ed., 1975: Three Voyages: René Laudonnière. Univ. Presses of Florida, Gainesville.

Benson, Adolph B., ed., 1964: Peter Kalm's Travels in North America: The English Version of 1770 (Vol. II). Dover Pubs., Inc., New York.

Bergen County Office of Cultural and Historic Affairs, 198/3: Bergen County Historic Sites Survey, Borough of Oakland. Bergen County Office of Cultural and Historic Affairs, Hackensack, NJ.

Beverley, Robert, 1705 [1947]: The History and Present State of Virginia, in Four Parts. Univ. No. Carolina Press, Chapel Hill.

Brinkhuizen, D.C. and A.T. Clason, eds., 1983: Some notes on recent and pre- and protohistoric fishing gear from northwestern Europe. Paleohistoria 25:7-53.

Brumbach, Hetty Jo, 1978: Middle Woodland Fishing Economics: The Upper Hudson River Drainage. Ph.D. dissertation, State Univ. of New York at Albany.

Brumbach, Hetty Jo, 1986: Anadromous Fish and Fishing: A Synthesis of Data from the Hudson River Drainage. Man in the Northeast 32:35-66.

Brydon, Norman F., 1974: The Passaic River - Past, Present, Future. Rutgers Univ. Press, New Brunswick, NJ.

Bushnell, David I., 1930: The five Monacan towns in Virginia, 1607. Smithson. Misc. Coll. 82(12).

Bushnell, David I., 1935: The Manahoac tribes in Virginia, 1608. Smithson. Misc. Coll. 94(8).

Chapman, James H., 1975: A Preliminary Report of Aboriginal Fish-Weirs in the Southeast. Unpublished MS, Georgia State University, 27 May 1975.

Chapman, Jefferson, 1985: Tellico Archaeology. Tennessee Valley Authority Publications in Anthropology No. 41, Univ. Tennessee Press, Knoxville.

Cobb, James E., 1978: Historic fish traps on the lower Holston River. Tenn. Anthropologist 3(1):31-58.

Connaway, John M., 1982: Sturdivant Fishweir. Southeastern Arch. 1(2):138-163.

De Yoe, Willard C., 1961: Looking Back in History - Bounty in Passaic River.

Dincauze, Dena F., 1973: Prehistoric occupation of the Charles River estuary: a paleogeographic study. Arch. Soc. Conn. Bull. 38:25-39.

Dincauze, Dena F., 1976: The Neville site : 8,000 years at Amoskeag, Manchester, New Hampshire. Peabody Museum of Archaeology and Ethnology, Harvard University, Cambridge, Mass.

Emery, K.O and R.L Edwards, 1966: Archeaological potential of the Atlantic continental shelf. Am. Antiq. 31(5):733-737.

Fisher, Charles, 1983: Social Organization and Change During the Early Horticultural Period in the Hudson River Valley. Ph.D. Dissertation, SUNY-Albany.

Flannery, Regina, 1939: An Analysis of Coastal Algonkian Culture. Catholic Univ. Of America Anthropological Series, #7. Catholic Univ. Of America, Press, Washington, D.C.

Funk, Robert E., 1983: The Northeastern U.S. In Ancient North Americans (J.D. Jennings, ed.). W.H. Freeman and Co., New York, 302-371.

Funkhouser, W.D. and William S. Webb, 1932: Archaeological survey of Kentucky. Univ. Kentucky, Reports in Arch. and Anthro. Vol. 2.

Godwin, Luke, 1988: Around the traps: a reappraisal of stone fishing weirs in northern New South Wales. Arch. of Oceania 23:49-59.

Gramly, Richard M. and Robert E. Funk, 1990: What is known and not known about the human occupation of the Northeastern United States until 10,000 B.P. Arch. of Eastern No. Am. 18:5-31.

Gravely, Richard P., Jr., 1973: Data Sheet for Proposed "Mini-park" on Smith River Adjacent to Frith Shopping Center on Route 220 South. (Unpub. MS).

Gravely, Richard P., Jr., n.d.: Private papers; copies in author's possession; provided by V. Steponaitis, Univ. North Carolina at Chapel Hill.

Hakluyt, Richard, 1907: The principal navigations, voigas, traffiques and discoueries of the English nation, Vol. G. Bishop & R. Newberies for C. Barker, London.

Hariot, Thomas, 1588 [1972]: "brief and true report of the new found land of Virginia..." Virginia: Four Personal Narratives. Arno Press, New York.

Heath, Dwight B., ed., 1963: A Journal of the Pilgirms at Plymouth: Mourt's Relation... Corinth Books, New York.

Heidenreich, Conrad E., 1971: Huronia: A History and Geography of the Huron Indians, 1600-1650. McLelland & Stewart, Toronto.

Hemmings, E. Thomas, 1970: Archaeological survey of the Trotters Shoals Reservoir area in South Carolina. Inst. of Arch. and Anthro., Univ. So. Carolina, Research Manuscript Series, vol. 3.

Hemmings, E. Thomas, 1973: Update on the Trotter's Shoals Reservoir. Notebook V(2), Inst. of Arch. and Anthro., Univ. So. Carolina.

Heusser, Albert H., 1923: Homes and Haunts of the Indians. Braen-Heusser, Paterson, N.J.

Holmes, W.H., 1896: Prehistoric textile art of eastern United States. Bur. Am. Ethnol., Thirteenth Annual Report.

Hubbert, Charles and Richard A. Wright, 1987: Lalakalka, the fishing place. J. Alabama Archaeol. 33(1).

Jenkins, Ned and Richard A. Krause, 1986: The Tom Bigby Watershed in Southeastern Prehistory. Univ. Alabama Press.

Johnson, Frederick, ed., 1949: The Boylston Street Fishweir II. Papers of the Robert S. Peabody Foundation for Archaeology, Vol IV, Number 1, Andover, Mass.

Johnson, Frederick, et al., 1942: The Boylston Street Fishweir. Papers of the Robert S. Peabody Foundation for Archaeology, Vol. 2, Andover, Mass.

Johnston, Richard B. and Kenneth A. Cassavoy, 1978: The fishweirs at Atherley Narrows, Ontario. Am. Antiq. 43(4):697-709.

Jones, Andrew K.G., 1986: Fish bone survival in the digestive systems of the pig, dog and man: some experiments. In Fish and Archaeology: Studies in Osteometry, Taphonomy, Seasonality and Fishing Methods. B.A.R. International Series 294, Oxford, pp. 53-61.

Jones, Charles C., 1973 [1873]: Antiquities of the Southern Indians, Particularly of the Georgia Tribes. Appleton and Co., New York.

Kraft, Herbert C., 1976: Archaeological and Historical Survey of the Proposed Right of Way for the Sanitary and Interceptor Sewer Lines, Borough of Oakland, Bergen County, New Jersey. Archaeological Research Center, Seton Hall University Museum, South Orange, NJ.

Kraft, Herbert C., 1984: The northern Lenape in prehistory and early colonial times. In The Lenape Indian - A Symposium. (H.C. Kraft, ed.) Seton Hall Univ. Museum, South Orange, N.J., 1-10.

Kraft, Herbert C. and Alan Mounier, 1982: The Archaic period in New Jersey (ca. 8000 B.P. - 1000 B.C.). In New Jersey's Archaeological Resources from the Paleo-Indian Period to the Present. (O. Chesler, ed.) Office of New Jersey Heritage, Trenton, 52-102.

Langdon, Steve J., Douglas Reger, and Christopher Wooley, 1986: Using Aerial Photographs to Locate Intertidal Stone Fishing Structures in the Prince of Wales Archipelago, Southeast Alaska. Alaska Division of Geological and Geophysical Surveys, Public Data File 86-9, Anchorage.

Larson, Lewis H., 1980: Aboriginal Subsistence Technology on the Southeast Coastal Plain During the Late Prehistoric Period. Univ. Presses of Florida, Gainesville.

Lee, R.B. and I. Devore, 1968: Man the Hunter. Aldine Press, Chicago.

Lenik, Edward J., 1985: The Archaeology of Wayne. Wayne Township Historical Commission, Wayne, N.J.

Lorant, Stefan, editor, 1946: The New World: The First Pictures of America. Duell, Sloan and Pearce, New York.

Louis Berger & Associates, Inc., 1987: Area B (28 Me 1-B): Archaeological Data Recovery, I-295, and Wetlands Area Interchange (Trenton Complex Archaeology: Report 8). Louis Berger & Associates, Inc., Cultural Resouce Group, East Orange, NJ.

Mahaffy, John J., 1978: A typology and functional analysis of fish weir configuration, R.L. Harris Reservoir. In Archaelogical Investigations in the R.L. Harris Reservoir (C. Hubbert, ed.), Appendix 5; MS on file, Office of Archaeological Research, Univ. Alabama.

Marshall, Sydne B., 1983: Cultural Resources Reconnaissance, Ramapo River -- Oakland, Pompton Lakes and Wayne, New Jersey. U.S. Army Corps of Engineers, New York Division.

McClane, A.J., ed., 1978: Field Guide to Freshwater Fishes of North America. Holt, Rinehart and Winston, New York.

Merriweather, Robert Lee, 1940: The expansion of South Carolina, 1729-1865. Southern Publication 194, Kingsport, Tenn.

Moorehead, Warren K., 1931: The Merrimack Archaeological Survey. Peabody Museum, Salem, MA.

Moss, Madonna L. and Jon M. Erlandson, 1990: Wood stake weirs and salmon fishing on the Northwest Coast: Evidence from southeast Alaska. Can. J. Arch. 14:143-158.

Myer, William E., 1928: Indian trails of the Southeast. Report of the Bureau of American Ethnology, 1919-24, pp. 727-857.

Peck, Rodney M., 1977: Stone fish traps in the Upper Pee Dee River. The Chesopiean 15(1-2):2-8.

Perlman, Stephen M., 1980: An optimum diet model, coastal variability, and hunter-gatherer behavior. In Advances in Archaeological Method and Theory, Vol. 3 (M.B. Schiffer, ed.). Academic Press, New York, 257-310.

Proctor, Mary A., 1930: The Indians of the Winnipesaukee and Pemigewasset Valleys. Towne & Robie, Pubs., Franklin, NH.

Rau, Charles, 1885: Prehistoric fishing in Europe and North America. Smithson Contributions to Knowledge 25:1-342.

Rights, Douglas L., 1929: A Voyage Down the Yadkin-Great Peedee River. Winston-Salem, NC.

Ritchie, William A., 1956: Prehistoric settlement patterns in northeastern North America. In Prehistoric Settlement Patterns in the New World (G.R. Willey, ed.). Wenner-Gren Foundation for Anthropological Research, Inc., New York, 72-80.

Ritchie, William A., 1980: The Archaeology of New York State (revised ed.). Harbor Hill Books, Harrison, N.Y.

Rogers, Robert Q., 1960: From Slooterdam to Fair Lawn. Thomas Jefferson Junior High School, Fair Lawn, N.J.

Rostlund, Erhard, 1952: Freshwater fish and fishing in native North America. Univ. Calif. Pubs. in Geog., Vol. 9.

Schalk, Randall F., 1977: The structure of an anadromous fish resource. In For Theory Building in Archaeology (L.R. Binford, ed.). Academic Press, San Francisco, 207-249.

Shields, Wayne F., 1967: The Chariton River fish trap. Missouri Hist. Rev. 61(4):489-496.

Skinner, Alanson and Max Schrabisch, 1913: A Preliminary Report of the Archaeological Survey of the State of New Jersey. MacCrellish and Quigley, Trenton.

Smullen, Edward L., 1921: Surveys Done for His Excy General Washington by Robert Erskine 1778-1779... (retracing).

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

Speck, Frank G., Royal B. Hassrick and Edmund S. Carpenter, 1946:Rappahannock taking devices: traps, hunting and fishing. Joint Pubs., Mus. of the Univ. Penn. and the Phila. Anthropol. Soc., No. 1.

Speth, John D. and K.A. Spielmann, 1983: Energy source, protein metabolism, and hunter-gatherer subsistence strategies. J. Anthropological Archaeology 2(1):1-31.

Strandberg, Carl H., 1962: Ancient Indian Fishtraps in the Potomac River. Photogrammetric Engineering 28(3):476-478.

Strandberg, Carl H. and Ray Tomlinson, 1969: Photoarchaeological analysis of Potomac River fish traps. Am. Antiq. 34:212-219.

Swanton, John R., 1946: The Indians of the southeastern United States. Bur. Am. Ethnol. Bulletin No. 137.

Swezey, Sean L. and Robert F. Heizer, 1984: Ritual regulation of anadromous fish resources in native California. In The Fishing Cultures of the World - Studies in Ethnology, Cultural Ecology and Folklore, Vol. 2. Akadémiai Kiadó, Budapest.

Thomas, Cyrus W., 1894: Report on the mound explorations of the Bureau of American Ethnology. Bur. Am. Ethnol., Twelfth Annual Report, 1890-1891, pp. 3-742.

Trigger, Bruce G., ed., 1978: Handbook of North American Indians, Vol. 15 - Northeast. Smithsonian Institution, Washington, D.C.

Van der Donck, Adriaen, 1656 [1968]: A Description of New Netherlands (ed. by Thomas O'Donnell). Syracuse Univ. Press, Syracuse, N.Y.

Wheeler, Alwyne and Andrew K.G. Jones, 1989: Cambridge Manuals in Archaeology: Fishes. Cambridge Univ. Press, New York.

Whyte, Thomas R., 1988: Fish and shellfish use in the Woodland Period on the Virginia Coast. J. Middle Atlantic Archaeol. 4:105-120.

Williams, Lorraine E. and Ronald A. Thomas, 1982: The Early/Middle Woodland period in New Jersey (ca. A.D. 1000-1600). In New Jersey's Archaeological Resources from the Paleo-Indian Period to the Present (O. Chesler, ed.). Office of New Jersey Heritage, Trenton, 103-138.

Williams, Samuel Cole, ed., 1927: Lieut. Henry Timberlake's Memoirs, 1756-1765; with Annotations, Introduction, and Index. Wautauga Press, Johnson City, Tenn.

Wintemberg, W.J., 1902: Report on a supposed aboriginal fish-weir in Burgess Lake, near Drumbo, Ontario. Ann. Archaeol. Report 1901:35-38, Appx. to Rep., Minister of Education, Ontario.

Woodward, C.R., 1941: Ploughs and Politicks - Charles Read of New Jersey and his Notes on Agriculture. Rutgers Univ. Press, New Brunswick, N.J.

al's home page