X Tutup
Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Apr 14;16(4):e0247643.
doi: 10.1371/journal.pone.0247643. eCollection 2021.

Archaeological science meets Māori knowledge to model pre-Columbian sweet potato (Ipomoea batatas) dispersal to Polynesia's southernmost habitable margins

Affiliations

Archaeological science meets Māori knowledge to model pre-Columbian sweet potato (Ipomoea batatas) dispersal to Polynesia's southernmost habitable margins

Ian G Barber et al. PLoS One. .

Abstract

Most scholars of the subject consider that a pre-Columbian transpacific transfer accounts for the historical role of American sweet potato Ipomoea batatas as the kūmara staple of Indigenous New Zealand/Aotearoa Māori in cooler southwestern Polynesia. Archaeologists have recorded evidence of ancient Polynesian I. batatas cultivation from warmer parts of generally temperate-climate Aotearoa, while assuming that the archipelago's traditional Murihiku region in southern South Island/Te Waipounamu was too cold to grow and store live Polynesian crops, including relatively hardy kūmara. However, archaeological pits in the form of seasonal Māori kūmara stores (rua kūmara) have been discovered unexpectedly at Pūrākaunui on eastern Murihuku's Otago coast, over 200 km south of the current Polynesian limit of record for premodern I. batatas production. Secure pit deposits that incorporate starch granules with I. batatas characteristics are radiocarbon-dated within the decadal range 1430-1460 CE at 95% probability in a Bayesian age model, about 150 years after Polynesians first settled Te Waipounamu. These archaeological data become relevant to a body of Māori oral history accounts and traditional knowledge (mātauranga) concerning southern kūmara, incorporating names, memories, landscape features and seemingly enigmatic references to an ancient Murihiku crop presence. Selected components of this lore are interpreted through comparative exegesis for correlation with archaeological science results in testable models of change. In a transfer and adaptation model, crop stores if not seasonal production technologies also were introduced from a warmer, agricultural Aotearoa region into dune microclimates of 15th-century coastal Otago to mitigate megafaunal loss, and perhaps to support Polynesia's southernmost residential chiefdom in its earliest phase. A crop loss model proposes that cooler seasonal temperatures of the post-1450 Little Ice Age and (or) political change constrained kūmara supply and storage options in Murihiku. The loss model allows for the disappearance of kūmara largely, but not entirely, as a traditional Otago crop presence in Māori social memory.

PubMed Disclaimer

Conflict of interest statement

The author has declared that no competing interests exist.

Figures

Fig 1
Fig 1. Location maps and archaeological site plan.
(A) Polynesia in the setting of eastern Oceania, after map modified from SVG file ‘World map blank without borders’, GNU Free Documentation License, CC BY-SA 4.0, adding island locations from public domain map data available from U.S. Geological Survey (USGS), National Geospatial Program. (B) New Zealand/Aotearoa after map data sourced from Land Information New Zealand (LINZ) Data Service licensed for reuse under CC BY 4.0 and locations in [28,30,33,36,66,74]. Traditional Murihiku incorporates Otago region to Waitaki River (cadastral border enclosing lighter fill), as well as the southern, modern Southland region (Fig 6.8 in [30]), [33]. Shown generally only are Canterbury region north of Waitaki River and the Ngāti Porou tribal area of eastern Te Ika-a-Māui. (C) Plan of Pūrākaunui archaeological pit complex, I44/21 S dune terrace. Elevation above mean sea level (DVD1958) is determined by DGPS.
Fig 2
Fig 2. Excavation plan for I44/21 S pit structures.
Pits coded P1-P4 (in bold, codes cross-referenced to Fig 1) are interpreted as subsurface components of semi-subterranean stores. All postholes recorded from the greater I44/21 S excavation area are represented. In P3, postholes from units H5 and H8 incorporating discrete starch-bearing fill are coded individually. The wall collapse along northern P3 side is an archaeological event capped by L2. Coded P3 excavation sections reference Figs 3 and 4; see also Figs 1 and S1 for further context, and S1 Table for unit descriptions.
Fig 3
Fig 3. Section images from excavated pits, I44/21 S (referenced from Fig 2).
Arrows indicate oblique angle of fill lenses and associated mollusc valve orientation. Unit codes follow S1 Table. (A) Northern edge of P2 truncating L2 beside external posthole (text over fill), image in perpendicular view; scale bar 10 cm. (B) Western P3 edge, unit corner and internal posthole (text mainly left of fill), oblique view; scale increment 5 cm. (C) Southeastern P4 edge, unit corner, oblique view; scale increment 20 cm.
Fig 4
Fig 4. Images of P3 excavation section, I44/21 S.
(Upper.) Main image, full section referenced from Fig 2, east face, perpendicular view. The locations of two dated marine samples (Paphies australis) are identified by 14C Lab numbers. Arrows indicate mollusc valve orientation in fill lenses (tape scale 10 mm increments). (Lower.) Detail of closed, articulated P. australis bivalve dated as Wk-37521 (right of tape scale at 550 mm increment), oblique view.
Fig 5
Fig 5. Bayesian chronological model for calibrated 14C ages by phase from I44/21 S incorporating Outlier [O:] analysis in OxCal v. 4.4 (after model details, code and data in S2 Text and S2 Table).
The model applies calibration curves SHCal20 and Marine20 [43,44], with the latter corrected for local reservoir variation using a weighted mean ΔR value from historic mollusc data (-162±28) updated for Marine20 (S2 Text). Calibration ranges for marine (green) and atmospheric (grey) determinations present prior probability distributions in lighter fill and modeled posterior distributions in darker fill. Posterior intervals are shown below individual histograms at 68% and 95% probabilities. Prior outlier probability was set at 0.05 (5%) for all determinations except Wk-37501 (prior probability 100%), as outlined in the text.
Fig 6
Fig 6
Starch granules from P3, unit H8 posthole fill (I44/21 S, panels A-C), and from reference I. batatas roots (D-F). Paired images in brightfield (left) and polarized light (right), with the hilum shown at the intersection of a birefringent extinction cross in polarization. Arrows point to nonround cavities and fissures at the hila of granules >13.2 μm. In granules of two panels (C, F), hila of archaeological and reference specimens present with distinctive y-shaped fissures reported previously for reference I. batatas (Fig 2a in [46], Fig 9e and 9f in [51]).

References

    1. Kirch PV, Green RC. Hawaiki, Ancestral Polynesia: An Essay in Historical Anthropology. Cambridge: Cambridge University Press; 2001.
    1. Kirch PV. On the Road of the Winds: An Archaeological History of the Pacific Islands before European Contact. Rev. & exp. ed. Oakland, CA: University of California Press; 2017.
    1. Kirch PV. The Wet and the Dry: Irrigation and Agricultural Intensification in Polynesia. Chicago and London: University of Chicago Press; 1994.
    1. Ballard C, Brown P, Bourke RM, Harwood T, editors. The Sweet Potato in Oceania: A Reappraisal. Sydney: University of Sydney; 2005.
    1. Clarke AC, Burtenshaw MK, McLenachan PA, Erickson DL, Penny D. Reconstructing the origins and dispersal of the Polynesian bottle gourd (Lagenaria siceraria). Mol Biol Evol. 2006;3: 893–900. 10.1093/molbev/msj092 - DOI - PubMed

Publication types

LinkOut - more resources

X Tutup