Arrival routes of first Americans uncertain—Response

See allHide authors and affiliations

Science  16 Mar 2018:
Vol. 359, Issue 6381, pp. 1225
DOI: 10.1126/science.aar8645

In our Perspective, we argue that the earliest Americans followed a coastal migration route. Potter et al. counter that an inland, ice-free corridor route was more likely. The genetic and archaeological evidence that Potter et al. discuss support both the timing and coastal route for the peopling of the Americas presented in our Perspective. Potter et al. are disappointed, however, that we did not also emphasize the possibility of an interior ice-free corridor migration and more tightly constrain colonization to after 16,000 years ago. This misses the main points of our Perspective: A Pleistocene coastal migration into the Americas has been bolstered tremendously by the discovery of early coastal and near coastal sites in the Americas, and more research, particularly underwater archaeology, is badly needed.

Potter et al. imply that in the most recent survey “archaeologists were divided, with many thinking both interior and coastal routes were used.” In fact, 86% of respondents selected “coastal migration,” compared with 65% for “interior passage migration” (1). This represents a sea change from 30 years ago when funding dollars and research efforts were funneled toward ice-free corridor models of New World colonization, resulting in decades of marginalization of research into early coastal migrations and adaptations (2).

Potter et al. underestimate the complexity of finding intact late Pleistocene coastal sites. Even in geological contexts where Pleistocene shorelines remain aboveground (such as in Alaska and British Columbia), coastal erosion and other forces tend to ravage or obscure early sites. In contrast to the sustained and intensive efforts to identify early sites in interior regions, the search for Pleistocene sites in coastal settings is just getting started. A growing number of such sites have been identified in places where systematic efforts to find late Pleistocene to early Holocene coastal sites have been made: the Pacific Northwest, the islands of Alta and Baja California, and coastal Peru and Chile (36).

Recent genomic evidence suggesting a migration into the Americas from Northeast Asia between ∼16,000 and 13,500 years ago does not preclude a coastal migration, and ages based on molecular clocks are approximations due to unpredictable variation in mutation rates. Potter et al. suggest that we argue for a migration earlier than 16,000 years ago based on the distribution of stemmed projectile points. We did argue that northeast Asian stemmed point technologies may support Native American origins sometime in the past 20,000 years—like genetic data—but that more work is needed to link these Asian technologies to similar but younger terminal Pleistocene toolkits in the Americas.

Potter et al. subscribe to a window of colonization between about 16,000 and 13,500 years ago, the latter half of which is almost certainly incorrect given the widely accepted 14,500-year-old occupation at Monte Verde (7). A coastal route was open by ∼17,000 years ago, whereas luminescence dates suggest that the ice-free corridor was open and viable by 15,000 to 14,000 years ago (8), or slightly later (9). The latter route would have allowed little time for human migrants to reach and establish residency at terminal Pleistocene sites in Oregon (10), Florida (11), Peru (4), Chile (7), and other far-flung regions.

The timing and viability of the ice-free corridor route remains uncertain, and arguments that the initial peopling of the Americas followed a terrestrial route hinge on its reconstruction. Questions remain about the coastal migration theory as well, and current evidence suggests that there likely were multiple dispersals, routes, and time frames for the peopling of the Americas.


Stay Connected to Science

Navigate This Article