Figure 1.
Direct dates on charred soybean seeds.
Black bars indicate 2σ range; white box indicates 1σ range of the calibrated dates. PLD9088: Shimoyakebe, Japan; MTC05837: Shimoyakebe, Japan; B289364: Dahecun, China; B25927: Pyeonggeodong, South Korea; TO8611: Okbang 1/9 (Nam River), South Korea; UCIAMS60750: Daundong, South Korea; TO8610: Three Kingdom, South Korea. B, Beta Analytic, USA; PLD, Paleo Labo Co., Ltd., Japan; MTC, Research Center of Nuclear Science and Technology, University of Tokyo, Japan; TO, University of Toronto Isotrace Laboratory, Canada; UCIAMS, University of California Irvine Keck Carbon Cycle AMS Facility, USA. All the materials dated are charred. Conventional dates were calibrated with Calib 6.0 using the Intcal 09 curve [8], [9], [10].
Figure 2.
Locations of Sites discussed in the text.
Figure 3.
Chronology relevant to the regions discussed in paper.
The Proto-Zhou period represents non-Shang culture in western Shaanxi that is contemporaneous with Late Shang centered in Henan.
Figure 4.
Hilar view of reference soybean.
A. Modern wild soybean (Accession No. IT209387, dark colored seed, left). B. two variations of a small-seeded domesticated soybean (Accession No. IT209387, light colored seeds, right) illustrating main characteristics. The seed coat covers the radicle so only its outline is visible. C. Commercial soybean (separate scale) with seed coat removed to expose the radicle and imprint of the hilum.
Figure 5.
Hilar view of charred Neolithic soybean seeds.
A. Dahecun. Specimen with seed coat attached showing a promiment hilum and faboid split; B. Jiahu. Most of the seed coat is absent exposing the bulbous radicle; C. SEM photograph of specimen from Dahecun with most of the seed coat missing but the hilum is still visible and the radicle is exposed; D. Pyeonggeodong. SEM photograph of soybean seed with patches of seed coat remaining; the bulbous form of the radicle is visible in the broken region at the top left of the image. Scale bar = 1 mm.
Figure 6.
Hilar view of charred Mumun soybean.
Daundong site. SEM photograph of seed with no seed coat; the radicle is at the top of the image while in the lower half is the faboid split in the hilar region. Scale bar = 1 mm.
Figure 7.
Hilar view of soybean seeds from Japan.
A. Seed from the Late Jomon Usujiri Shogakko site; the entire seed coat is missing. B. Seed with well-preserved seed coat from the Middle Jomon occupation of the Shimoyakebe site. Scale bar = 1 mm.
Figure 8.
SEM photo illustrating details of a mature seed coat cross-section of a charred seed from Dahecun.
Detail of seed coat surface on Dahecun seed (A) and modern wild seed (B) illustrating the typical honeycomb pattern of the bloom on seed coat exterior that shows that the seed had a dull luster. Such blooms are characteristic of both wild and domesticated soybean and variations of the bloom deposits determine whether a seed is shiny or dull.
Figure 9.
Scatter plot of length vs. width of archaeological soybean.
The density ellipse (red) represents the 90 percent confidence limit of the regression line for modern wild soybean.
Table 1.
Principal Domestication Related Traits (DRT) Distinguishing Wild from Domesticated Soybean.
Figure 10.
Box plots of soybean seed length, length by width, length × width, and length × width × thickness.
Numbers indicate seed numbers measured. Measurements from different sites for each period are combined. The top, bottom and line through the middle of the box correspond to the 75th percentile (top quartile), 25th percentile (bottom quartile) and 50th percentile (median) respectively. The whiskers on the bottom extend from the 10th percentile (bottom decile) and top 90th percentile (top decile). Solid boxes represent means. Samples represented by a box have one or only a few specimens. Only the range is published for the Late Jomon sample from Kyushu.
Figure 11.
Comparison of shape and size of late soybean populations.