Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 90 Introduction Animals are ubiquitous in Algonquian oral traditions in the Northeast, often depicted as transformational beings with special qualities and powers, and with the spirit of the slain animal seen as determining the success of future hunts (Betts et al. 2012, Hornborg 2008, Rand 1971, Tanner 1979). Special qualities often extended to the bones of the animals and influenced how they were processed. Here we attempt to distinguish between symbolic and utilitarian treatments of bone. More specifically, we identify ways in which different species were treated differently in different contexts. This approach is particularly appropriate for our research on Machias Bay (Fig. 1) where we are working with the Passamaquoddy Tribal Historic Preservation Office to better understand the area’s unusual concentration of petroglyphs (Hedden 1996, 2004; Soctomah 2009). While petroglyphs may automatically be classified as symbolic, their intended use may have been quite practical. In either case, we are interested in why some animals are portrayed and others are not in different contexts. Shell-midden archaeology provides an exceptional opportunity for addressing these kinds of questions. We present 3 cases from recent excavations and analyses of two eastern Maine shell middens, including: (1) special disposal of gray seal bone (Ingraham et al. 2016), (2) retention of mandibles and maxilla of the extinct sea mink, and (3) inclusion of mandibles and skulls in domestic or ritual structures. The findings are preliminary but are presented here to highlight testable patterns as well as the opportunities and difficulties involved. Background Ethnographic examples of specialized animal processing for the purpose of respecting the spirit of the animal are widespread and varied. These include burning bones to keep them from being eaten by dogs (Sanger 2003:31, Thwaites 1896:210), returning aquatic animal bones to the river or sea (Sanger 2003:31, Soctomah 2002:171, Speck 1935a:23), and placing bones of terrestrial animals in trees or containers or on platforms (Preston 1964:144; Smith 2011:80;Tanner 1979:153, 172). Ritual activities involving animal remains also include divination practices such as scapulimancy, which entails reading the cracks of a burned scapula or other bones to predict good hunting locations (Speck 1935b:127, Tanner 1979:108). Selected bones can serve as hunting talismans or personal amulets (Speck 1935b:30, Tanner 1979:141). Animal skulls and claws may be retained as parts of animal skin on clothing, in or as an element of medicine bundles, or for badges of office (Fox and Molto 1994, Willoughby 1905). The case for special treatment of animal bone is strengthened by its broad cultural occurrence, providing a general context for behaviors that contrast with more utilitarian expectations in modern industrial societies (Trigger 2006:526, Viveiros de Castro 1998). All of these practices have counterparts among the Wabanaki tribes of Maine and the Canadian Maritimes, including the Penobscot, Passamaquoddy, Maliseet, and Micmac (Wabanaki Program of the American Friends Service Committee 2002) which supports their applicability here (Ingraham et al. 2016, Le Clercq 1910:226, Sanger 2003, Smith 1954:36, Soctomah 2002:171, Speck 1935a, Whitehead 1991:17). The general practice of special treatment of animal bones is abundantly documented, and special treatments of some species are well recorded, but particular treatments of many bone elements will have to be uncovered from the archaeological record. With so broad a record and so many potential uses, it follows that bone symbolism may be a prominent part of well-preserved faunal assemblages in coastal Maritime Culture Patterns and Animal Symbolism in Eastern Maine Brian S. Robinson1,† and A. Sky Heller1,* Abstract - Coastal and maritime environments provide a whole series of environmental and geographical factors that are integrated into our understanding of past cultural landscapes. These include both advantageous factors such as enhanced bone preservation associated with shell, as well as more deleterious factors such as site loss from sea-level rise. Good preservation can provide opportunities to explore archaeologically the ritual aspects of human–animal relationships—distinguished from more utilitarian aspects of subsistence processing—for which substantial records of oral traditions and beliefs exist. Here we focus on examples from recent excavations on Machias Bay and Frenchman Bay in Maine, spanning 4000 years. North American East Coast Shell Midden Research Journal of the North Atlantic 15677 South Steven Hall, Orono, ME 04469. †Deceased. *Corresponding author - a.sky.heller@gmail.com. 2017 Special Volume 10:90–104 Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 91 shell middens. However, with the explicit search for such practices, there are all too many suspicious associations and possibilities. The problem lies in finding multiple cases with sufficiently convincing contexts to correctly identify practices that range from individual variation to regional traditions. One challenge in identifying the symbolic aspects of faunal remains lies in understanding how systematic or pervasive different symbolic usages might be. Speck (1935b: 217) recorded for the Montagnais or Innu of Quebec that removing the hide of a beaver with a skinning tool made of bear bone was a “religious obligation”; this practice could represent a powerful regional cultural pattern. Sanger (2003:29) proposed that processing bone to avoid disrespectful scavenging by dogs may have operated at the scale of “a deep-seated, interior/coastal divide that extends to the basic cosmological relationship between humans and prey.” At an even larger scale, it has been proposed for the Gulf of Maine Archaic tradition of New England and Maritime Canada that the absence of chipped-stone projectile points, combined with cremation of bone tools in primary burial rituals, might represent widespread religious obligations regarding the appropriate material for killing game (Robinson and Ort 2011). Utilitarian explanations may fail to explain such prevalent aspects of the cultural record. Cultural norms can be powerful motivators, while the degree to which the ideal is followed in practice may be highly variable (Trigger 1995:454). Father Le Jeune (Thwaites 1896:210) recorded in 1634 the complex practices of the Montagnais of Quebec, with regard to bone disposal: The Savages do not throw to the dogs the bones of female Beavers and Porcupines,—at least, certain specified bones; in short, they are very careful that the dogs do not eat any bones of birds and of other animals which are taken in the net, otherwise they will take no more except with incomparable difficulties. Yet they make a thousand exceptions to this rule, for it does not matter if the vertebrae or rump of these animals be given to the dogs, but the rest must be thrown into the fire. Yet, as to the Beaver which has been taken in a trap, it is best to throw its bones into a river. It is remarkable how they gather and collect these bones, and preserve them with so much care. Further complicating things, while the dichotomy between utilitarian (e.g.,subsistence, tool production) and more symbolic (or ritual) uses of bone can serve as an important distinction in grouping causal factors, it can be quite inappropriate to define mutually exclusive domains where they are in fact diverse and overlapping. For similar reasons, we do not advocate for a particular theoretical position, certainly not between the materialist/idealist dichotomies of the past, but rather accept the essential and often complementary tensions between perspectives that investigate both general and historical processes (Sassaman and Holly 2011:5, VanPool and VanPool 2003:3). Animal bones can obviously represent both subsistence and ritual contexts (Trigger 1995:451), but ferreting out different relationships requires recognition of multiple contexts, sufficient distinctiveness in the archaeological record, the means to test between alternate scenarios at different scales, and recognition of when these scenarios may be overlapping and interdependent. One way of identifying ritual usages is to contrast species representations between domestic and mortuary contexts (Betts et al. 2012, Holt 1996). In the modern dichotomy between utility and symbolism, intentional inclusion within mortuary contexts almost automatically qualifies animal remains as ritual or symbolic, even if intended as food for the dead. Among examples from Newfoundland and Canadian Maritimes are the abundance of bird and mammal bones buried in the Late Archaic period Port au Choix cemetery (Tuck 1976), the prominence of bird skulls in graves of the recent Beothuk (Kristensen 2010:43), and inclusion of fossil and recent shark teeth in mortuary contexts (Betts et al. 2012). This paper focuses on another domain for symbolic usage: identifying repetitive patterns in domestic contexts that are structured differently from those of biological, subsistence, or manufacturing requirements. Despite potential overlap among these analytical categories, identifying specific contexts of meaning requires distinguishing amongst the myriad natural and cultural influences on bone deposition and preservation (Lyman 1994). Maximizing historical and processual considerations is a good thing. The present effort is simply to identify contrasting structural relationships and contexts set within general definitions of subsistence and more culturally defined uses of bone. For the past 7 years, the University of Maine has excavated the Holmes Point West shell midden in Machias, ME, USA (Fig. 1), collaborating with the Passamaquoddy Tribe to better understand the abundant Machias Bay Petroglyphs. Prior excavations at the Holmes Point West site were conducted by Robert MacKay in 1973, yielding a large (if unscreened) faunal sample that was analyzed by Robert Ingraham (2011) for his Master’s thesis in the Climate Change Institute. A pattern of seal processing was identified in which the remains of large gray seal (Halichoerus Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 92 grypus) were significantly dominated by the left temporal bone (and the distinctive auditory bulla) of the skull. This finding was placed in the context of oral traditions that call for the return of bones of marine animals to the sea (Ingraham et al. 2016). A second faunal pattern at Holmes Point West, recognized at other shell middens in coastal Maine, is the almost exclusive representation of the extinct sea mink (Neovision macrodon formerly Mustela macrodon; Mead et al. 2000) by cranial parts with few post-cranial bones. The recent Holmes Point West excavations contributed a cluster of 3 mostly complete sea mink maxillae or upper jaws in a welldefined feature, with 2 other maxillae located within 2.5 m. Excavations were also conducted at the Waterside site in Sorrento, ME, USA, in relation to Sky Heller’s dissertation research at the University of Maine. Heller’s research focuses on investigating changes in fish ecology at a time when swordfish disappeared from the archaeological record in the Gulf of Maine, ca. 3800 radiocarbon years BP. The Waterside site was previously excavated by John Rowe in the 1938 and stands as one of the few remaining Late Archaic period shell middens on the Gulf of Maine. It thus provides important ecological and stratigraphic information for what is now called the Moorehead phase and its relationship to the Susquehanna tradition and the later Ceramic period (Rowe 1940). Rowe donated the site to the Archaeological Conservancy in 1999 (Bangor Daily News 1999). Our excavations in 2013 were directed to relocating stratigraphy from Rowe’s excavations that hinted at a possible Late Archaic period structure. Evidence for this structure and an apparent concentration of mammal mandibles were recovered during removal of soil columns for fine screening. In each of these three case studies, different criteria must be weighed in evaluating symbolic patterning relative to other potential explanations. Each is pronounced enough to distinguish it from patterns of direct refuse disposal. Each is the subject of further testing and continued search for alternate explanations. Together they provide examples of symbolic activities, evidence of cosmology and identity, and potential links to the animal symbolism of oral traditions that would be very difficult to make without the preservation afforded by shell middens. Holmes Point West Site and Machias Bay The Holmes Point West site is located on the east side of Machias Bay in the town of Machiasport, ME, USA (Fig. 1). The current excavations are being conducted as a field school funded by a Maine Academic Prominence Initiative grant (MAPI; Newman and Robinson 2011) intended for the salvage of eroding shell middens and for incorporating the interests and community members of Maine’s Wabanaki tribes. Our involvement in Machias developed from the 2006 return of the Birch Point petroglyph site on Machias bay to the Passamaquoddy through cooperation of the Passamaquoddy tribe and Maine Coast Heritage Trust (2006) and subsequent development of a management plan to protect and develop cultural and educational programs around the petroglyph landscape (Soctomah 2009). The Figure 1. Site plan of Holmes Point West, and (inset) site locations mentioned in text along the central coast of Maine, USA: (1) Turner Farm, (2) Waterside, (3) Great Spruce Island, and (4) Holmes Point West. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 93 association of these petroglyphs—the largest concentration of such images on the east coast of North America, spanning an estimated 2500 years (Hedden 1996, 2004)—with shell middens and good bone preservation provides a remarkable window on animal/ human relationships through time. The present case studies are part of this larger ongoing project, involving multiple research projects (graduate and undergraduate) and preservation efforts. None of the shell middens on Machias Bay are undisturbed. They are frequently located on undefended Pleistocene marine sediments (as at Holmes Point West) which, along with high tidal ranges (up to 4 m), has resulted in extensive erosion of sites around the Bay. Artifact collecting at these sites, once a more popular pastime, has also taken its toll. Although relatively shallow (20–40 cm) and partly disturbed, the site was unplowed, and bioturbation remains one of the major disturbances on extant surfaces. Progressive erosion has resulted in the loss of earlier shell strata, with the migration of more recent shell deposits over earlier occupation areas that were formerly behind the midden. Radiocarbon dates on pre-shell midden features are as old as 2165 ± 15 BP (ISGS A2005, processed by Karine Tache, on residue from a pseudo-scallop shell potsherd from Feature 8). Radiocarbon dates on Features 21 and 28 (Fig. 1), respectively, representing the initiation of the central shell deposit, are 860 ± 30 BP (Beta-408150, Cal AD 1150 to 1250 at 2σ) and 740 ± 30 BP (Beta- 408153, Cal AD 1250 to 1290 at 2σ). Thus, the major shell deposit and preserved bone date to between ca. 1200 AD and 1700 AD, into the colonial period. This period overlaps with the proposed age of the major petroglyph ledge at the Birch Point site, located directly across Machias Bay. The latest petroglyphs at Birch Point include a single-masted 17th-century ship with sail and multiple crosses (Hedden 2004:342). Holmes Point: The left ear bone of the gray seal Excavations by the University of Maine in 1973 (Fig. 1) yielded a substantial faunal sample that has been described in detail (Ingraham 2011, Ingraham et al. 2016). The assemblage was dominated by seals with a minimum number of individuals (MNI) of 11 gray seals (Halichoerus grypus) and 11 harbor seals (Phoca vitulina). The next highest MNIs are 4 moose (Alces alces) and 4 razorbill (Alca torda; Ingraham et al. 2016:Table 1). Among the issues represented by these species are the widespread Algonquin accounts, including those of both the Passamaquoddy (Soctomah 2002:171) and Penobscot (Speck 1935a:23), of the disposal of the bones of aquatic animals back into the water, effectively removing them from the archaeological record. How could this apply if the most abundant species at the site are marine mammals? Does the oral tradition not apply at Holmes Point? Or, can we discern some of the local complexity behind the general oral tradition, as in the remarkable quotation from Le Jeune above? The most numerous element for both gray and harbor seal—providing the high minimal counts above—is the temporal bone of the skull (Fig. 2; Ingraham et al. 2016:table 2). This bone is often the most abundant seal element found in sites around the Gulf of Maine (Spiess and Lewis 2001:69) and in northern Europe (Storå 2001:Paper V:13), which should automatically raise a red flag. It is a highly distinctive bone (identifiable in small fragments) and durable, especially the massive mastoid process. High visibility and durability are powerful attributes in archaeology that should be approached with caution. It is therefore important to thoroughly address taphonomic issues that could account for the patterns in other ways (Harper 1999, Ingraham et al. 2016:95). This is especially true because the sample was Figure 2. Left temporal bone with auditory bulla from a gray seal (PN 2940.2, excavated in 2013). Side view looks through the auditory canal. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 94 collected without screening and also used for class instruction. At the same time, large sample sizes of seal elements are unusual in Maine, and the dominance of skull fragments does not represent preferred meat patterns, meriting attention. In the present case, six of eleven left gray seal bulla came from a single large feature (Feature 73-3) that allowed us to evaluate taphonomic and recovery characteristics (Ingraham et al. 2016:93). The feature was an isolated shell-filled pit set back 3 m from the edge of the shell midden in which bone was well preserved with four species of fish including sturgeon, cod, blue fish, and American eel. A large moose bone shaft was radiocarbon dated to 430 ± 30 BP (Beta–408151, Cal AD 1430 to 1485 at 2σ), although upper levels included early 17th-century contact-period artifacts. Analyses of preservation and collection procedures suggest that neither would account for the pattern of large and well-preserved seal bones, although small bone fragments were certainly missed. In the case of gray seal, 11 individuals are represented by 11 left and only 2 right temporal bones, representing a statistically significant dominance of the left side (Ingraham et al. 2016). Sidedness is not a factor of either visibility or durability. Does it represent intentional selection? Theodore White (1956:401) observed that it is not uncommon to find significant differences in sidedness, noting that in “certain groups the parents and grandparents of the man and wife customarily received specific elements of the carcass, such as the left front leg.” But the temporal bone of the skull is not a meaty bone, making habitual division of a carcass an unlikely explanation. The skull may have been retained as a container for the brain for use in brain tanning of hides (Harper 1999:121). A concerted effort to locate other skull elements did not yield any, and using the skull as a container would not account for sidedness patterning. Alternatively, the well-documented practice of retaining and protecting specific bones from each animal hunted may have incorporated sidedness (Preston 1964:144; Smith 2011:80; Tanner 1979:153, 172). An example exists among the Wabanaki, although its origins are disputed, where the left hind foot of the moose was sometimes retained for medicinal purposes (Ganong 1908:382, Merrill 1916:263). A dominance of right dog tarsals is recorded (six right calcanea, 3 with cut marks, and 4 right naviculars) at the Indian Island site in the Sheepscot River estuary of central coastal Maine, USA, suggesting the possibility of cultural selection factors (Spiess et al. 2006:165). Spiess and Lewis (2001) regularly searched for the significance of sidedness at the Turner Farm site in Maine, as noted below. Testing with 95% probability means that one in twenty cases may be accidental, emphasizing the need for sufficient sample sizes and contextual information. In contrast to the temporal bone, there are marked absences among other gray seal bones. Although post cranial bones are generally difficult to distinguish between gray and harbor seal, other bones are large, durable, and easily identifiable, especially the mandible. The 1973 excavations produced 21 fragments of gray seal temporal bones, but no mandibles. This is in contrast with harbor seal, with 22 fragments of temporal bone, mostly from the right side, and 11 mandible fragments, suggesting different treatments for gray and harbor seal (Ingraham et al. 2016:Table 2). One possible explanation for the different treatment of the two seal species is dissimilar butchery practices related to a difference in size. Oral traditions suggest that the bones of marine mammals were “left for the tide to take back to the ocean” (Soctomah 2002:171). If smaller harbor seals were brought back to the habitation site for butchery, this may have resulted in a greater proportion of post-cranial remains deposited locally. Larger gray seals, however, may have been butchered remotely, possibly directly on the shore, decreasing this possibility. The left temporal bone, however, seems to have been singled out and retained, conforming to ethnographic accounts of special treatment of animal bones. The gray seal may have multiple treatments, like the beaver reported by Le Jeune, and the contrast between disposal offsite and the retention of specific bones potentially enhances the visibility or the pattern. Ingraham identified the pattern in the 1973 excavation sample, and we get to test it with ongoing excavations and analysis. Holmes Point: Facial bones of the sea mink The sea mink was first described as an extinct species by Prentiss (1903) from a single archaeological skull fragment recovered in Brooklin, ME, USA. The type specimen consisted of the maxilla, nasals, and zygoma, lacking the posterior brain case (Prentiss 1903:888). Since that time, there has been debate about whether the sea mink is a separate species (Loomis 1911, Mead et al. 2000, Prentiss 1903) or a subspecies (Hardy 1903, Manville 1966) of the modern American mink (Neovision vision, formerly Mustela vison). The most recent and most comprehensive analysis favored a separate species (Mead et al. 2000). Also of considerable interest are interpretations of the distribution of sea mink. Black et al. (1998) suggest that sea mink recovered in New Brunswick were traded or brought from Maine, based on interior occurrences and association with non-local lithics. This is an important issue even at Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 95 and provide preliminary observations here as well as a more detailed account of Feature 28. In the first 2 seasons (2008 and 2009), a small number of sea mink mandibles were recovered. In 2010, two fragments of maxilla were recovered in the screen from the NE quadrant of unit N28 E19 (Provenience Number 2557), but no feature was recognized. In 2012, Joshua W. Desrosier and Mandi Curtis excavated the adjacent unit N28E20, and on June 25, Josh carefully exposed 7 teeth and maxilla fragments that were photographed or plotted in situ. All of the students had been instructed in identifying sea mink and seal bones but it is a credit to Josh’s attentiveness and record keeping that the small bones were recorded in situ. This proved a valuable record because it was later determined that the tightly clustered specimens (designated Feature 28) originated from 3 nearly complete maxillae or facial bones from 2 large (Nos. 1 and 2; Fig. 3) and 1 smaller mink (No. 3; Fig. 3). The area of concentration (area 2557; Fig. 3) contained a total of 17 mink teeth and maxilla fragments from level 6 of this quadrant of the feature (including the 7 that were in situ), with only 2 or 3 other bone fragments. The two fragments discovered the previous year (PN 1559) were reattached to minks No. 2 and 3. One additional maxilla fragment from the left side of mink No. 1 was found in situ in 2014 (PN 3335.2; Fig. 4). Once reassembled, all fragments were accounted for by the 3 mink maxillae, including both the left and right sides of all 3 specimens. The larger specimens have been identified as sea mink due to their size and morphology. Mink No. sites within the proposed range within Maine when only a few selected elements are found. The Turner Farm site in Penobscot Bay (Fig. 1; Bourque 1995) produced the largest well-documented assemblage of mink bones, spanning 5000 years of occupation, with 1004 elements of which 2% to 6% are thought to be American mink and the remainder sea mink (Spiess and Lewis 2001:74). Of interest here is that bone elements from all parts of the body are present for both the Archaic and Ceramic periods, and that an analysis of left and right sides suggests “some sort of differential human treatment of right versus left mandibles” in some periods but not consistently over time with a total of 132 left mandibles and 137 right (Spiess and Lewis 2001:75). In 1909, the Amherst Biological Expedition collected faunal remains from shell heaps along the Maine coast. Flagg Island, in Casco Bay produced what was interpreted as “10 upper and 34 lower jaws of males, and 2 upper and 11 lower jaws of females” identified as sea mink (Loomis 1911:227). Given that these were excavated in one week, it is not expected that all elements would have been collected equally, but it is noted that “[e]very skull has the brain case broken and lost” while the “facial portion of each skull is, however, pretty much intact…” (Loomis 1911:227). Other sites produced only 2 or three mandibles. At Great Spruce Island in Englishman Bay, just west of Machias Bay, Sanger and Chase (1983) recorded 12 bones of the extinct sea mink, 11 of which were mandibles, 9 of which were from the left side representing another possible case of selection that is discussed further below. This site is of particular interest due to its proximity to Machias Bay and because the excavators were able to distinguish clear gravel house floors behind the shell midden. The house areas were represented by cranial elements alone, while the shell dumping area had both cranial and post cranial bone (Sanger and Chase 1983). At Holmes Point West, the 1973 excavations produced two mink bones, both of which were right mandibles (Ingraham et al.2016). Faunal analysis from our excavations between 2008 and 2014 is ongoing, but we were alert for sea mink Figure 3, Three mink skulls (1–3) and four teeth of the angler fish (bottom row) from Feature 28. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 96 ern quadrants of Feature 28 and retained all the soil for fine-screening. Although we have not identified mink phalanges from these samples, 11 long angler teeth (Lophius sp.; Fig. 4) were recovered from a 50 m x 1.5 m area overlying the feature, only one of which came from a fine-mesh screen, assuring comparability across the site. These teeth might have acted like tinkling cones that were often attached to the feet of medicine bundles (Harrison 1986, Willoughby 1905:638). Although faunal remains are still being processed, angler teeth are thus far limited to the area of Feature 28, and no other angler skeletal elements have been identified. If they had been found in groups of 5 among the sea mink maxillae, a good case might be made for medicine bundles, but they were scattered over the feature, apparently above the level of the sea mink bones. They seem to be associated with the feature, but not yet in a recognizable spatial pattern. With excavation of the northern quadrants of Feature 28, the structure of the feature itself became more apparent. One stone tool—a complete ground stone celt—was recovered from the NE quadrant (Fig. 3). Large flakes and rocks appear to represent a covering. A beaver (Castor canadensis) pelvis in the NW quadrant was found in the same level as the mink maxilla. A portion of the pelvis (PN 3331.3) was radiocarbon dated to 860 ± 30 BP. Other studies are ongoing. Samuel Hatch and Emily Blackwood, who excavated portions of the northern half of the feature, worked with Andrea Nurse (University of Maine, Climate Change Institute, Orono, ME, USA) to process soil samples for pollen and phytolyths. Andrew Heller prepared a micromorphology column for microscopic examination of the feature and events that produced it. Kendra Bird is investigating spatial relationships of the flaking debris and other materials in and around the feature. While it would be very valuable if another similar feature was found, it is remarkable how informative a single feature with complex contexts can be. We do not need to test the statistical significance of three mink facial bones to identify whether they are associated by chance (although statistical analysis of other contents will be useful). Rather, multiple lines of evidence identify this as an event, a moment in time associated with complex activities. It is entirely possible that sea mink do not represent subsistence remains at all at Holmes Point West, as suggested for New Brunswick (Black et al. 1998). Do the 12 maxillae recovered on Flagg Island in Casco Bay represent similar cultural practices of a broad coastal pattern (Loomis 1911)? Certainly the dominance of mandibles within domestic structures at nearby 3 is less clear; it is close in size to a modern American mink, but the shape of the infraorbital foramen is more similar to sea mink. By good fortune and accidents of timing, Feature 28 was neatly quartered and taken out over a period of 4 years. The first year we missed it altogether. The majority of mink bones came out in the last week of the second year, which initiated research on the cultural uses of mink facial bones. Possible uses include talismans and clothing attachments as well as retention of skulls in skin sacks and bundles. Otterskin medicine bundles as well as mink and other animals were commonly used throughout the Great Lakes region, especially associated with Midewiwin or Grand Lodge Medicine Society of the Ojibwa, within which each individual had their own bag (Densmore 1929:93, Hoffman 1891). A mink bag retaining the skull of the mink is illustrated in Densmore (1929:93, Plate 36) for the Chippewa, west of Lake Superior. Facial bones of an otter have been interpreted as a medicine bundle from a reputed shaman burial in Ontario, Canada, dated to ca. 900 AD, calibrated (Fox and Molto 1994:36). Among Great Lakes groups, the otter is the pre-eminant medicine animal associated with a wide variety of contents that are identifiable archaeologically (Claassen 2014:129, Harper 1999:358). The mink maxillae may or may not be from medicine bundles, and we sought alternate ways of testing this possibility. An occasional attribute of medicine bundles is that the phalanges and claws are left as part of the bag. To search for these associated parts, in 2013 and 2014, we excavated the 2 north- Figure 4. Plan of Feature 28 that was excavated in quadrants. Provenience numbers (PN) are provided for selected artifacts. Rocks are in black. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 97 Great Spruce island is related somehow (Sanger and Chase 1983). Notably, although the faunal analysis at Holmes Point West is not finished, sea mink remains are strongly dominated by cranial elements, but the mandibles and maxilla have not been found together. If supported with increased sample sizes, these are consistent retention and disposal practices that are good candidates for strongly maintained cultural practices. With Feature 28, we are beginning to discern some of the complex associations of the sea mink. The Waterside Site and Mammal Mandibles The Waterside Site in Sorrento, ME, USA, is one of the rare Late Archaic period shell middens on the Gulf of Maine, dating to about 4000 cal BP. The earliest excavations at the site were conducted by John Rowe in 1938 and 1940 when the property belonged to his family. Only a limited portion was excavated, preserving valuable samples of this small site for modern analyses. Rowe did a particularly good job of describing the stratigraphy in his 1940 publication. He later became a well-known Peruvian archaeologist but retained the property in Sorrento and donated the site to the Archaeological Conservancy in 1999 as a permanent archaeological preserve. Correlation of our work with that of Rowe was enhanced when Ann Surprenant recognized that Rowe’s excavation units were still visible on the surface, so close to the paved road that it was difficult to believe that they had not been disturbed. The Waterside site is bedrock defended, but it is still eroding. Rowe’s accurate map and the visibility of his excavations on the surface allowed us to accurately measure the amount of erosion that occurred between 1940 and 2013 (Fig. 5), which we calculated to be up to 2 m of site loss, including about two rows of the 1940 excavation units. The Archaeological Conservancy was interested in further work at the site due to site loss. Sky Heller’s research on Late Archaic period marine ecology required minimal Figure 5. Excavation plan of the Waterside from 1940, superimposed over modern topographic map with a 10-cm contour interval. Note depressions from original excavation units. Units with heavy boundaries were re–excavated in 2013, with column samples numbered 1–4. Plan by Andrew Heller. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 98 The top few inches of the humus and pebbles are practically sterile, containing no shells, bones, or specimens [artifacts]. As one goes deeper, the humus gets richer, the pebbles scarcer, and in the central and eastern sections are vast quantities of animal bones, with frequent specimens: plummets, hammerstones, celts, bone points, slate points, adzes, and swordfish sword. The rich humusbones- and-pebbles layer rests directly on the yellow glacial clay which forms the bottom of the shell heap. Our excavations corroborated the description of a high density of bone at the base of the stratum, where we found a swordfish vertebra lying partly in contact with the underlying yellow brown subsoil. This deep, bone-rich, shell-free stratum was more reminiscent of a deep pit filling than a surface deposit, which was supported by the deep shell layer recorded on the eastern edge of unit 2N/1W, described as: excavation to recover soil columns for fine screening. We removed backfill from 2 of Rowe’s units (Fig. 5), in locations with Moorehead phase shell-midden and non-shell deposits. We then removed four (25 cm square) columns for fine screening, a total area of 0.25 m2. We did not expect to find evidence of animal bone symbolism in the very small area of undisturbed strata that we excavated. That we seem to have found it makes it intuitively more convincing, but the work is necessarily preliminary. Rowe identified 2 periods of occupation at the Waterside site, an upper Ceramic-period shell midden and a more complex lower Moorehead-phase occupation, although these terms were not in use at that time. The Moorehead-phase stratigraphy was quite different in different areas, represented by the 2 units that we re-excavated in 2013, Rowes’s 3N/2W and 2N/1W (Fig. 5). We provide our profile only for the eastern unit, 2N/1W (Fig. 6). Rowe’s (1940:7) description of the non-shell Moorehead-phase layer applies to the more westerly unit 3N/2W as well as to our stratum G in unit 2N/1W (Fig. 6): Figure 6. Waterside excavations (2013) correlated with the Rowe’s unit 2N 1W, and with North and West walls of the extension associated with column removal. Positions of mandibles are shown. Rocks R1, R2, and R3 are the same ones plotted by Rowe. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 99 Excavations through the “whole-shell” yielded swordfish ribs and the left and right paired, partlyarticulated, mandibles of a moose that retained 3 incisors (PN 259; Fig. 7). Both sides of the mandible were broken off 15 cm from the anterior end of the jaw, and the broken posterior ends were battered or flaked and polished, perhaps by handling or bag polish. This find resembles broken and paired mandibles from Woodland period burials (Fox and Molto 1994:32), but as of yet we have seen no comparable specimens from moose. Directly at the base of the whole shell, just to the side of the moose mandibles, was a nearly complete deer mandible with the lower edge broken away (PN 261; Fig. 7). While clearing a pocket between the rocks on the north wall at about the same time, we exposed a complete right bobcat (Lynx rufus) mandible, missing the canine tooth. Correspondence from John Rowe indicates that he recovered a lynx (or bobcat) mandible in his excavation of the 2N units (records of the Abbe Museum, Bar Harbor, ME, USA). At Waterside, it is the more general selection of large-mammal mandibles—including scarce (bobcat) and unusual (articulated left and right … a very clean, loosely packed layer of whole and large broken shells at the bottom, resting on some fire-blackened rocks. The way in which the layers overlap makes it clear that these shell layers antedate the old surface formed elsewhere by the humus and pebble layer, and they contain no pottery. (Rowe 1940:7) During our excavation of Rowe’s unit 2N/1W, we were easily able to identify his stratigraphy, even to the extent of identifying specific rocks in the lower right corner of the north wall profile (Rocks 1, 2, and 3 in Fig. 6; and Rowe 1940:Plate XIV). It was clear that Rowe’s “humus and pebbles” (our stratum G) overlay the “whole and large broken shells” (our stratum D), an arrangement that resembled the filling of semi-subterranean house structures from the Ceramic period of eastern Maine (Sanger 2010). While collecting soil columns for fine screening, we extended this unit to the east and encountered more loose shell underlain by dense boulders. In our new excavations, it was apparent that the boulders projected higher into the “whole shell” layer than apparent in Rowe’s excavations, that some boulders lay in direct contact with those below, that the lower boulders lay mostly on black soil beneath the shell midden (that we interpret as the original top soil) although some intruded into the subsoil, and that the “whole-shell” (stratum D) was deposited directly on and between the boulders (Fig. 6). The whole-shell layer appears to be draped over the boulders. There was no indication of a pit below the boulders. This arrangement resembles Ceramic period house pits surrounded by rocks (thought to support rafters) and with peripheral deposits of shell around the rocks (Sanger 2010:27). Although there is currently only a trench across the Waterside feature, if this interpretation is correct, then our 25-cmsquare soil columns landed directly on the peripheral stone deposit of a house structure. Figure 7. Three mandibles removed from the black soil below the rock pile (PN 212 and 261) and from the overlying whole shell layer (PN 259) at the Waterside site. From top to bottom they are moose, bobcat, and deer. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 100 distal halves of moose) mandibles—and their placement within a proposed architectural feature that is proposed as a case of animal-bone symbolism. Similar associations occur in the Ceramic period on Great Spruce Island, where cranial elements and mandibles were reported to be concentrated in Ceramic-period houses (Sanger and Chase 1983). At Waterside, bone symbolism in an occupation context complements evidence of animal-bone symbolism in mortuary contexts of the Moorehead burial tradition (Claassen 2015), with less support from the direct historical approach. Discussion and Connections The 3 case studies presented above represent special processing and use of faunal remains that are proposed to relate to different symbolic functions and culturally appropriate use of animal bone. Each is distinguished by a different set of contexts and a different set of test implications, and each, in effect, requires development of culturally specific middlerange theory (Trigger 1995), ferreting out sometimes idiosyncratic impacts on the material record that can be identified archaeologically. The left temporal bone of the seal was distinguished as a statistically significant pattern from a whole-site faunal sample, spanning hundreds of years. In this case, it is the high frequency of one bone element contrasted with the low number of other diagnostic elements, such as the mandible, that suggests both a pattern of offsite disposal and cultural selection, both practices recorded ethnographically and in oral traditions. The challenges of identification involve sufficient sample sizes of elements from a relatively stable cultural practice, while testing involves replication of the results in other samples and sites, and better definition of specialized contexts. Better identification of postcranial bones might provide more highly structured differences in the processing of gray and harbor seals, a study that is underway. The selection of cranial elements of the sea mink was recognized at other sites (Loomis 2011, Sanger and Chase 1983). The new evidence from the Holmes Point involves a concentration of facial bones within a complex feature. The use of mink and mustelid facial bones has direct ethnographic analogues in the use of medicine bags for which middle-range hypotheses may contribute support for symbolic function of this element. The complexity of the feature itself was only recognized because of the effort to refine the context of the sea mink concentration relative to surrounding activities, based on ongoing studies by Andrew Heller, Kendra Bird, and others. Interpretations have changed and developed. The feature may be a covering on an activity floor rather than a pit or a hearth, with rather unique formational processes that will help identify similar cases. In the meantime, gleaning the contexts of this feature and analyzing historical contexts continues. The Waterside site stratigraphy published over 70 years ago by John Rowe hinted at the presence of some sort of domestic or ritual structure, and our recent excavations provide further support for this. The association of multiple, unusual mandibles with a possible architectural stone embankment is unique for the northeastern Archaic period but has analogies in the Ceramic period (Sanger 2010, Sanger and Chase 1983). House structures provide a key to numerous aspects of social and ritual organization, justifying the effort of detailed spatial excavation and analysis (Hrynick and Betts 2014). Although specific historical analogies may be weaker due to greater time depth at the Waterside site, the broader pattern of human/animal relationships and specialized processing and treatment of animal bone may well be of great time depth. This pattern may serve as a broad cosmological process with demonstrable impact on the archaeological record that contrasts with more utilitarian explanations for faunal distributions. Aspects of human/animal relations have been proposed to be more broadly Amerindian, and still more broadly, animistic (Claassen 1914, Viveiros de Castro 1998), but the variations are not universal and expressions of particular cultures can be both diverse and remarkably stable, among signatures of world view, culture, and identity (Betts et al 2012:624, Robinson and Ort 2012). These cultural processes, in turn, impact many other aspects of faunal analysis. We have not proved specific meanings for any of the case studies, but have identified species-specific symbolic contexts that can be tested and developed. These are necessary steps in the recognition of symbolic factors that may greatly influence the structure of faunal samples from occupation sites. The effort expended on such enquiries is correlated with values placed on them. The research in Machias was initiated as part of the Passamaquoddy Petroglyph Project. The association has strongly influenced: (1) the topics of the research, (2) the shared experience gained by involvement of Native and non-Native students and scholars, and (3) the explicit goal of finding connections between past and the present, topics elaborated on below. These are characteristics of an increasing number of archaeological projects, and it is useful to affirm that, in this case, a broadening of Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 101 interests and values enhanced the scientific aspects of archaeology by increasing the importance of the research and the range of factors that were investigated. First, the topic of this paper and the emphasis on symbolic uses of bone was to a large degree initiated by a search for symbolic contexts associated with the concentration of petroglyphs, and, in turn, on the importance of the petroglyphs to the Passamaquoddy. The petroglyphs concentrated on Machias Bay are links to a proud Passamaquoddy cultural heritage and identity. This was demonstrated during traditional ceremonies in recognition of the transfer of the “Picture Rocks” site. The transfer event included Passamaquoddy elders, a new generation of grade-school Passamaquoddy students as well as other interested individuals and comprised over 200 people. (Soctomah 2009) Thus, while shell middens are traditionally of great importance to archaeology, this significance is enhanced and paralleled by their modern cultural importance and efforts to understand this long-lived tradition. Second, the success of a field school depends on the interest and attentiveness of the students, and the enthusiasm generated by the experience. Fieldwork is a continuous development of new insights and interests. Field schools conducted at the Holmes Point West site included a tour of the major petroglyph site by Donald Soctomah communicating the immediacy of spiritual and cultural links for the Passamaquoddy. The shaman and animal figures in stone provide possible meanings for the fragmentary animal bones in the shell midden. We emphasize careful watch for particular faunal patterns as we learn them, and the day Josh Derosier uncovered 3 sea mink (in situ!), it was already part of a broader story. The week before the discovery, we took our field trip to the Passamaquoddy Tribal Museum in Indian Township, where Donald Soctomah and field-school participants Natalie and Cassandra Dana served as guides. These tours help form connections between archaeological materials and human activities. For example, something as “mundane” as fire-cracked rock, when encountered in action at an active sweat lodge, can take on a whole new significance for the student. The Tribal Museum houses a 6.4-m (21-ft) birch-bark canoe that was built in 2011 (Soctomah 2014) and launched at both Passamaquoddy reservations at Indian Township and Pleasant Point so that all the Passamaquoddy could use their Native craft. We were pleased, indeed, at the invitation to carry the canoe from the museum to nearby Grand Falls Flowage, where Donald took the stern and most of us had our first experience in a birch-bark canoe. The importance of participation presupposes the third important impact of the collaboration, linking the past to the present. Among the many cultural items in the Museum are etched birch-bark canoes and containers made by Tomah Joseph and his son Sabattus Tomah. Tomah Joe and Sabattus are the great grandfathers of Natalie and Cassandra Dana who are, in turn, board members of the Maluhsihikon Petroglyph Foundation and were part of the Machias field schools from 2008 to 2014. Sabattus Tomah spoke with ethnographer Nicholas Smith on numerous occasions in 1953 and 1954. Historian Micah Pawling, Donald Soctomah, and Brian Robinson are now working with Nick Smith to compile the accounts of these conversations. Sabattus Tomah was a tribal story teller, and his stories of the white weasel are among widespread accounts of powerful otters, weasels, and mink that peopled the oral traditions of the Wabanaki, providing windows on animal symbolism that we hope to find archaeologically. Sabattus Tomah also related an account of origins of the petroglyphs at Machias Bay, bringing traditional knowledge to the present. Donald Soctomah’s great grandfather, John Soctomah, was a seasonal resident of Machiasport and an important informant for Fannie Hardy Eckstorm’s (1978) research on place names, which are in turn major signposts on the cultural landscape (Basso 1996, Soctomah 2004). We were fortunate to interview Frank Foster, who remembered John Soctomah, and Wayne Renshaw, who lived next door to John Soctomah’s summer residence, providing surprising details of their relationship with the town. After Maine Coast Heritage Trust helped the Passamaquoddy acquire the major petroglyph site in Machiasport, the Trust went on to develop their first cultural heritage-based land holdings, not as an archaeological monument, but as a heritage tradition in which modern participation is seen as a key to preservation (Deirdre Whitehead, Maine Coast Heritage Trust, East Machias, ME, 2015 personal comm.). These last aspects of cultural collaboration may not all directly bear on the topic of this paper, except that much of the research would not have occurred were it not for the combined goals and efforts of those involved. Such collaborations are a part of many projects, often combining prehistoric and historic archaeology (Leone 2003) with a variety of current interests. Each project has to identify particular goals and potentialities, and how they fit into general enquiries and cultural values. The shell middens are eroding at an alarming rate. It is clear that if the loss of cultural heritage through erosion Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 102 Claassen, C. 2015. Beliefs and Rituals in Archaic Eastern North America: An Interpretive Guide. The University of Alabama Press, Tuscaloosa, AL, USA. 408 pp. Densmore, F. 1929. Chippewa customs. Smithsonian Institution Bureau of American Ethnology Bulletin 96, Washington, DC, USA. Eckstorm, F.H. 1978. Indian Place Names of the Penobscot Valley and the Maine Coast. First published 1941. University of Maine at Orono Press, Orono, ME, USA. 272 pp. Fox, W.A., and J.E. Molto. 1994 The shaman of Long Point. Ontario Archaeology 57:23–44. Ganong, W.F (Ed.) 1908. The Description and Natural History of the Coasts of North America (Acadia). by Denys, N. 1672. The Chaplain Society, Toronto, ON, Canada. 687 pp. Hardy, M. 1903. The extinct mink from the Maine shell heaps. Forest and Stream 6l:125. Harper, R. K. 1999. To render the God of the water propitious: Hunting and human–animal relations in the Northeast Woodlands. Ph.D. Dissertation. Paper AAI9956473. Available online at: http://digitalcommons. uconn.edu/dissertations/AAI9956473. Accessed 23 May 2016. Harrison, J. 1986. “He heard something laugh": Otter imagery in the Midéwiwin. Bulletin of the Detroit Institute of Arts 62:46–53. Hedden, M. 1996. 3500 years of shamanism in Maine rock art. Pp. 7–24, In C.H. Faulkner (Ed.). Rock Art of the Eastern Woodlands.American Rock Art Research Association. Occasional Paper 2, San Jose, CA, USA. 136 pp. Hedden, M. 2004. Passamaquoddy shamanism and rock art in Machias Bay, Maine. Pp 319–343, In C. Diaz- Granados and J.R. Duncan (Eds.). The Rock Art of Eastern North America. University of Alabama Press, Tuscaloosa, AL, USA. 456 pp. Hoffman, W.J. 1891. The Mide'wiwin or "Grand Medicine Society" of the Ojibwa. Pp. 143–300, In Seventh Annual Report of the Bureau of Ethnology to the Secretary of the Smithsonian Institution, 1885–1886. Government Printing Office, Washington, DC, USA. 409 pp. Holt, J.Z. 1996. Beyond optimization: Alternative ways of examining animal exploitation. World Archaeology 28:89–109. Hornborg, A.C. 2008. Mi’kmaq Landscapes: From Animism to Sacred Ecology. Ashgate Publishing Limited, Aldershot, UK. 214 pp. Hrynick, M.G., and M.W. Betts. 2014. Identifying ritual structures in the archaeological record: A Maritime Woodland period sweathouse from Nova Scotia, Canada. Journal of Anthropological Archaeology 35:92–105. Ingraham, R. 2011. Specialized taphonomies from an eastern Maine shell midden: Faunal analysis from site 62-8, Holmes Point West, Machias, Maine. Master’s Thesis. University of Maine, Orono, ME, USA. 111 pp. Ingraham, R.C., B.S. Robinson, K.D. Sobolik, and A.S. Heller. 2016. “Left for the tide to take back”: Specialized processing of seals on Machias Bay, Maine. Journal of Island and Coastal Archaeology 11(1):89–106. and uncontrolled digging of shell middens is to be minimized, it is going to have to come from shared interests and efforts. Acknowledgments This report provides case studies of elements from excavations that have involved numerous people, especially students and visiting scholars at the University of Maine field schools at Machias, and volunteers at the Waterside site. The field school program was funded by a grant from the Maine Academic Prominence Initiative (MAPI) to Brian Robinson and Lisa Neuman that inspired avenues of research development. The Machias work represents longterm collaboration with the Passamaquoddy Tribal Historic Preservation Office, the efforts of Maine Coast Heritage Trust, and landowners Robert and Kenneth Brack and families. The Bracks have been anonymous landowners in previous reports, and it is a pleasure to thank them for their enormous hospitality and for supporting so much of the preservation effort. Excavation of the Waterside site was conducted with the permission and encouragement of the Archaeological Conservancy. One of the case studies was the work of Robert Ingraham, whose recent paper we cite heavily. In the body of the report, we have named ongoing research by students and colleagues. Figures and supporting research were contributed by Andrew Heller and Kendra Bird. This paper was modified from ones presented at symposia chaired by Gabe Hrynick and Matthew Betts (AENA), as well as by Donald Holly and Christopher Wolf (CAA). Others copiously mentioned in the text but not to be missed here include Donald Soctomah, Nicholas Smith, Natalie Dana, Cassandra Dana, and Deirdre Whitehead. Ann Surprenant, Micah Pawling, and Donald Soctomah participated in research and provided comments on this paper. Cheryl Claassen, Ross Harper, and Gabe Hrynick provided helpful reviews. Shortcomings are our own. Literature Cited Bangor Daily News. 1999. Frenchman Bay Red Paint site preserved. October 25, p. B1. Basso, K. 1996. Wisdom Sits in Places: Landscape and Language Among the Western Apache. University of New Mexico Press, Albuquerque, NM, USA. 191 pp. Betts, M.W., S.E. Blair, and D.W. Black, 2012. Perspectivism, mortuary symbolism, and human–shark relationships on the Maritime Peninsula. American Antiquity 77:621–645. Black, D.W., J.E. Reading, and H.G. Savage. 1998. Archaeological records of the extinct Sea Mink, Mustela macrodon (Carnivora: Mustelidae), from Canada. Canadian Field-Naturalist 112:45–49. Bourque, B.J. 1995. Diversity and Complexity in Prehistoric Maritime Societies: A Gulf of Maine Perspective. Plenum Press, New York, NY, USA. 414 pp. Claassen, C. 2014. Hunting ritual, trapping meaning, gathering offerings. Pp. 125–160, In T.M. Peres (Ed.). Trends and Traditions in Southeastern Zooarchaeology. University Press of Florida Gainesville, FL, USA. 226 pp. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 103 Sanger, D. 2010. Semi-subterranean houses in the Ceramic Period along the coast of Maine. The Maine Archaeologial Society Bulletin 50:23–45. Sanger, D., and T. Chase. 1983. An introduction to the Roque Island Group, Maine. Paper presented to the Canadian Archaeological Association, April 1983. Manuscript on file at Anthropology Department, University of Maine, Orono, ME, USA. Sassaman, K.E., and D.H. Holly. 2011. Transformative hunter–gatherer archaeology in North America. Pp. 1–13, In K.E. Sassaman and D.H. Holly Jr. (Eds.). Hunter–Gatherer Archaeology as Historical Process. Amerind Studies in Archaeology, University of Arizona Press, Tucson, AZ, USA. 352 pp. Smith, N.N. 1954. Wabanaki dances. Bulletin of the Massachusetts Archaeological Society 16:29–36. Smith, N.N. 2011. Three Hundred Years in Thirty: Memoir of Transition with the Cree Indians of Lake Mistassini. Polar Bear and Company, Solon, ME, USA. 272 pp. Soctomah, D. 2002. Passamaquoddy at the Turn of the Century 1890–1920: Tribal life and times in Maine and New Brunswick. Passamaquoddy Tribe of Indian Township, Princeton, ME, USA. 177 pp. Soctomah, D. 2004. Landscapes, legends, and language of the Passamaquoddy People: An interactive journey in the land of the Passamaquoddy. CD, Skicin Records and Passamaquoddy Historic Preservation Office, Pleasant Point, ME, USA. Soctomah, D. (Ed.) 2009. Petroglyph management plan for Machias Bay: Protection and documenting the uniqueness of Machias Bay. Report to the National Park Service, Tribal Historic Preservation Office –Passamaquoddy Tribe, Pleasant Point Reservation, ME, USA. Soctomah, D. 2014. Passamaquoddy birchbark canoe. A production made possible by a National Historic Preservation Grant. Available online at http://www. youtube.com/watch?v=46LoBsZZwok. Accessed 1 September 2015. Speck, F.G. 1935a. Penobscot tales and religious beliefs. The Journal of American Folklore 48:1–107. Speck, F.G. 1935b. Naskapi; The Savage Hunters of the Labrador Peninsula. University of Oklahoma Press, Norman, OK, USA. 272 pp. Spiess, A.E., and R.A. Lewis. 2001. The Turner Farm Fauna: 5000 Years of Hunting and Fishing in Penobscot Bay, Maine. Occasional Publications in Maine Archaeology (11). Maine State Museum, Augusta, ME, USA. 177 pp. Spiess, A., K. Sobolik, D. Crader, J. Mosher, and D. Wilson. 2006. Cod, clams, and deer: The food remains from Indiantown Island. Archaeology of Eastern North America 34:141–187. Storå, J. 2001. Reading Bones: Stone Age Hunters and Seals in the Baltic. Stockholm Studies in Archaeology 21. 253 pp. Tanner, A. 1979. Bringing Home Animals: Religious Ideology and Mode of Production of the Mistassini Cree Hunters. St. Martin’s Press, New York, NY, USA. 233 pp. Kristensen, T.J. 2010. L’anse Aux Meadows (EjAv-01): An archaeological and ethnohistorical investigation of bird use during the Recent Indian Period in Newfoundland and Labrador. Master’s Thesis. Memorial University, St. John's, NL, Canada. 169 pp. Le Clercq, C. 1910. New Relation of Gaspesia, With the Customs and Religion of the Gaspesian Indians, W.F. Ganong (Ed.). The Champlain Society, totonto, ON, Canada. Leone, M. 2003. The origins of questions in historical archaeology. Pp. 17–22, In T.L. Vanpool and C.S. VanPool (Eds.). Essential Tensions in Archaeological Method and Theory. The University of Utah Press, Salt Lake City, UT, USA. 197 pp. Loomis, F.B. 1911. A new mink from the shell heaps of Maine. American Journal of Science 183:227–229. Lyman, R.L. 1994. Vertebrate Taphonomy. Cambridge University Press, Cambridge, UK. 552 pp. Maine Coast Heritage Trust. 2006. 3000-year-old petroglyph site regained by Passamaquoddy Tribe. Newsletters of the Maine Coast Heritage Trust. Available online at http://www.mcht.org/news/2006/10/3000_ year_old_petroglyph_site_1.shtml. Accessed 1 September 2015. Manville, R.H. 1966. The extinct sea mink, with taxonomic notes. Proceedings of the US National Museum 122:1–12. Mead, L I., A.E. Spiess, and K.D. Sobolik. 2000. Skeleton of extinct North American sea mink (Mustela macrodon). Quaternary Research 53:247–262. Merrill, S. 1916. The Moose Book: Facts and Stories from Northern Forests. E.P. Dutton and Company, New York, NY, USA. 496 pp. Neuman, L., and B.S. Robinson. 2011. Revealing shell midden legacies in coastal Maine: Building connections to local indigenous communities through a new kind of archaeological field school. Paper presented at the 110th Annual Meeting of the American Anthropological Association, Montréal, QC, Canada. Prentiss, D W. 1903. Description of an extinct mink from the shell-heaps of the Maine coast. Proceedings of the United States National Museum 26:887–888. Washington, DC, USA. Preston, R.J. 1964. Ritual hangings: An aboriginal “survival” in a northern North American trapping community. Man 64:142–144. Rand, S.T. 1971. Legends of the Micmacs. Johnson Reprint, New York, NY, USA. 317 pp. Robinson, B.S., and J.C. Ort. 2011. Paleoindian and Archaic Period traditions: Particular explanations from New England. Pp. 209–226, In K.E. Sassaman and D.H. Holly Jr. (Eds.). Hunter-Gatherer Archaeology as Historical Process. Amerind Studies in Archaeology, University of Arizona Press, Tucson, AZ, USA. 352 pp. Rowe, J. 1940. Excavations in the Waterside shell heap, Frenchman's Bay, Maine. The Excavators’ Club, Cambridge, MA, USA. 20 pp. Sanger, D. 2003. Who lived in pre-European Maine? A cosmology approach to social patterning on the landscape. Northeast Anthropology 66:29–39. Journal of the North Atlantic B.S. Robinson and A.S. Heller 2017 Special Volume 10 104 Thwaites, R.G. (Ed.). 1896. Relation of what occurred in New France in the year 1634, by Paul LeJeune. Pp. 5–32, In R.G. Thwaites (Ed.). Jesuit Relations and Allied Documents of the Jesuit Missionaries in New France: Quebec 1610–1791. Vol. 6. Burrows Brothers, Cleveland, OH, USA. 324 pp. Trigger, B.G. 1995. Expanding middle-range theory. Antiquity 69:449–458. Trigger, B.G. 2006. A History of Archaeological Thought. Second Edition. Cambridge University Press, New York, NY, USA. 710 pp. Tuck, J.A. 1976. Ancient People of Port au Choix: The Excavation of an Archaic Indian Cemetery in Newfoundland. Newfoundland Social and Economic Studies 17. Institute of Social and Economic Research, Memorial University of Newfoundland, St. John's, NL, Canada. 261 pp. VanPool, C.S., and T.L. Vanpool. 2003. Introduction: Method, theory, and the essential tension. Pp. 1–4, In T.L. Vanpool and C.S. VanPool (Eds.). Essential Tensions in Archaeological Method and Theory. The University of Utah Press, Salt Lake City, UT, USA. 197 pp. Viveiros de Castro, E. 1998. Cosmological deixis and Amerindian perspectivism. The Journal of the Royal Anthropological Institute 4:469–488. Wabanaki Program of the American Friends Service Committee. 2002. The Wabanakis of Maine and the Maritimes: A Resource Book about Penobscot, Passamaquoddy, Maliseet, Micmac, and Abenaki Indians. Wabanaki Program of the American Friends Service Committee, Philadelphia, PA, USA. White, T. E. 1956. The study of osteological materials in the Plains. American Antiquity 21:401–404. Whitehead, R.H. 1991. The Old Man Told Us. Nimbus Publishing Ltd., Halifax, NS, Canada. 385 pp. Willoughby, C.C. 1905. A few ethnological specimens collected by Lewis and Clark. American Anthropologist, New Series 7:633–641.