Late Pleistocene/Early Holocene sites in the montane forests of New Guinea yield early record of cassowary hunting and egg harvesting

Abstract

How early human foragers impacted insular forests is a topic with implications across multiple disciplines, including resource management. Paradoxically, terminal Pleistocene and Early Holocene impacts of foraging communities have been characterized as both extreme-as in debates over human-driven faunal extinctions-and minimal compared to later landscape transformations by farmers and herders. We investigated how rainforest hunter-gatherers managed resources in montane New Guinea and present some of the earliest documentation of Late Pleistocene through mid-Holocene exploitation of cassowaries (Aves: Casuariidae). Worldwide, most insular ratites were extirpated by the Late Holocene, following human arrivals, including elephant birds of Madagascar (Aepyornithidae) and moa of Aotearoa/New Zealand (Dinornithiformes)-icons of anthropogenic island devastation. Cassowaries are exceptional, however, with populations persisting in New Guinea and Australia. Little is known of past human exploitation and what factors contributed to their survival. We present a method for inferring past human interaction with mega-avifauna via analysis of microstructural features of archaeological eggshell. We then contextualize cassowary hunting and egg harvesting by montane foragers and discuss the implications of human exploitation. Our data suggest cassowary egg harvesting may have been more common than the harvesting of adults. Furthermore, our analysis of cassowary eggshell microstructural variation reveals a distinct pattern of harvesting eggs in late ontogenetic stages. Harvesting eggs in later stages of embryonic growth may reflect human dietary preferences and foraging seasonality, but the observed pattern also supports the possibility that-as early as the Late Pleistocene-people were collecting eggs in order to hatch and rear cassowary chicks.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 57 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.