Alaska Journal of Anthropology vol. 15, nos. 1&2 (2017) 7 THE MAGOUN CLAM GARDEN NEAR SITKA, ALASKA: NICHE CONSTRUCTION THEORY MEETS TRADITIONAL ECOLOGICAL KNOWLEDGE, BUT WHAT ABOUT THE RISKS OF SHELLFISH TOXICITY  Madonna L. Moss Department of Anthropology, University of Oregon, Eugene, OR, 97403; mmoss@uoregon.edu Hannah P. Wellman Department of Anthropology, University of Oregon, Eugene, OR, 97403; hpw@uoregon.edu ABSTRACT Located within the Magoun Islands State Marine Park, the Magoun Clam Garden (49-SIT-921) is, to our knowledge, the irst clam garden conirmed on the ground in Alaska. Although many “clam gardens” or “clam terraces” constructed by First Nations have been identiied in coastal British Co- lumbia, this is the irst that we can attribute to Alaska Natives, in this case, to Tlingit ancestors. he site was identiied by Madonna Moss, University of Oregon, and Aaron Bean, Sitka Tribe of Alaska, during archaeological survey in July–August 2011. In this paper we review some of the recent theo- retical approaches embedded in the study of these features and urge more explicit use of such theoreti- cal resources. We conclude that although niche construction theory (NCT), resource intensiication, resource management and conservation, and traditional ecological knowledge (TEK) provide useful frameworks for understanding these kinds of sites, only traditional ecological knowledge, and indig- enous cultural knowledge more generally, have recognized the health risks of shellish consumption caused by paralytic shellish poisoning (PSP). We suggest that studies of clam gardens must take into consideration such risks, which undoubtedly varied over time and across space. SITKA SOUND Along the shorelines of Sitka Sound, approximately ifty precontact archaeological sites have been identiied (Alaska Heritage Resource Survey [AHRS] site records and Jay Kinsman, pers. comm. 2011). hese include vil- lages, shell middens, ishing weirs and rock alignments, sites with depressions, garden rows, culturally modiied trees (CMTs), petroglyphs, camps, subsistence areas, and canoe haul-outs. he sites are located on state, federal, municipal, and private lands. Known sites comprise just a partial sample of the record. For example, we know Native Americans have occupied southeast Alaska for more than 12,000 years (Dixon 2013), but as yet no sites older than 2600 bp have been found in the Sitka Sound region. Moss’s interest in Sitka Sound archaeology began in 1982 when she was stationed in Sitka as an archaeologist working for the Chatham Area of the Tongass National Forest. In a recent review of the archaeological record of southeast Alaska conducted for the Herring Synthesis project (Moss et al. 2010, 2011; hornton et al. 2010a, 2010b), it was remarkable to discover how little archaeo- logical research has been conducted on precontact sites in and around Sitka, the center of commercial herring ish- ing today. Although some test excavations have occurred (e.g., Campen et al. 1992; Davis 1985; Erlandson et al. 1990), none of these reports have been published. Sitka National Historical Park has been completely inventoried, but the park only encompasses about 110 acres (Hunt 2010). Within park boundaries, eight precontact sites with