Keywords: zooarchaeology, genetics, species identification, environmental archaeology, Northwest Coast

Sturgeon (Acipenser spp.) have long been an important resource for people living in several parts of the world, including the Northwest Coast of North America. Two sturgeon species occur on the Oregon coast, white (A. transmontanus) and green (A. medirostris), but morphological similarities between the species have generally prevented osteological identification beyond genus. This lack of species identification hinders evaluation of potential changes in sturgeon range or distribution through time, human food preference or encounter rates, and possible size changes in archaeological sturgeon. We analyzed 1770 sturgeon remains from the Par-Tee site (35CLT20), northern Oregon coast and performed ancient DNA species identification on 30 elements to test morphological distinguishability between species. Both green and white sturgeon were harvested at the site, but identification to species based on morphology is problematic. Sturgeon were an important fish at several sites in the Northwest Coast and at Par-Tee were a component of a broad-based fishery focused on a range of large and small species. Our study emphasizes the importance of ancient DNA analysis of zooarchaeological assemblages to help address important questions about species composition and abundance.

In a 2017 article, Holen and colleagues reported evidence for a 130 000-year-old archaeological site in California. Acceptance of the site would overturn current understanding of global human migrations. The authors here consider Holen et al.'s conclusions through critical evaluation of their replicative experiments. Drawing on best practice in experimental archaeology, and paying particular attention to the authors' chain of inference, Magnani et al. suggest that to argue convincingly for an early human presence at the Cer-utti Mastodon site, Holen et al. must improve their analogical foundations, test alternative hypotheses, increase experimental control and quantify their results.