Supporting Information

Files in this Data Supplement:

Fig. 3. Assemblage of manioc (Manihot esculenta) starch from Casita de Piedra, flake chopper 52/44, ≈5600 cal BP. The granules range in size from 12.0 to 15.0 mm, with a mean length of 14.0 mm, and demonstrate the smooth, transparent ("shiny") surfaces, multiple pressure facets, centric, open hila, and fissures characteristic of manioc (17). (Scale bar: 10 mm.)

Fig. 4. Examples of maize (Zea mays) starch from Casita de Piedra from grinding-stone base 64/18, ≈4200 cal BP (A), and from chopper 76/2, ≈4200 cal BP (B). Granules are irregular rounded to polygonal in shape, with uneven surface, centric hila, and deep transverse to Y-shaped fissures. Average diameter of these and other maize granules from the site is 18.5 mm. (Scale bar: 10 mm.)

Fig. 5. Fabaceae (Legume family) starch from Casita de Piedra from edge-ground cobble 87/33, ≈5600 cal BP (A), and from edge-ground cobble 94/3, ≈6100 cal BP (B). Oval granules with ragged longitudinal fissures are characteristic among several members of the Fabaceae, including Phaseolus spp. Taxonomic identification of these granules to species may be possible with more comparative collecting and analysis. (Scale bar: 10 mm.)

Fig. 6. Assemblage of native yam (Dioscorea spp.) starch from Trapiche chopper 147/10, ≈4300 cal BP, showing the diversity of morphotypes. At least two different Dioscorea species were processed on this tool, possibly more. (A) Another example of D. cf. urophylla (see Fig. 2D). (C-E) Grains were photographed under cross-polarized light and clearly show the eccentric hilum (without fissures) and arms of the extinction cross which describe the cunate shape of the grains. (Scale bar: 10 mm.)

Fig. 7. Examples of maize starch from Trapiche from chopper 147/10, ≈4300 cal BP (A), and from chopper 153/1, ≈4900 cal BP (B and C). Grains are polygonal in shape with uneven surfaces, centric hila, transverse to stellate fissure, and average 19.0 mm in diameter. (Scale bar: 10 mm.)

Fig. 8. Additional examples of maize starch from Hornito, ≈7000 cal BP from wedge 77-1 (A), and from chopper C24 (B). Granules are similar in morphology to those seen at Casita de Piedra and Trapiche, and average 17.8 mm in diameter. (Scale bar: 10 mm.)

Fig. 9. Additional examples of maize starch from Cueva de los Ladrones from handstone CL-82b, ≈7800 cal BP (A), from handstone CL-68/1, ≈7500 cal BP (B and C), and a cluster of maize starch from the same tool (D). Maize granules were irregular polygonal in shape, with centric hila and transverse to Y-shaped fissures. Average size was 17.3 mm. (Scale bar: 10 mm.)

Table 2. Starch granules recovered from tools

Radiocarbon calibrations were made by using the OxCal 4.0 program (16). Numerals represent the number of granules identified from each taxon on a tool. Numerals in parentheses represent tentative identifications.

SI Materials and Methods

Starch grains are produced by many plant taxa as a mechanism for energy storage, and are most commonly found in seeds and tuberous roots. This stored energy is used during periods of regrowth after dormancy (such as a dry season), or germination (1). Botanists have long recognized that starch grain morphology is distinctive for different plant taxa (2, 3), but the archaeological application of starch grain analysis to questions of residue analysis and paleoethnobotany has only recently developed (4-11). Researchers have discovered that starch grains are often preserved in the microcrevices of stone tools used for plant processing. By extracting these grains and examining their microscopic morphology, diagnostic taxa can be identified, and provide data regarding artifact use, processing techniques, subsistence patterns, and plant domestication.

Methods used in this study followed those developed by various researchers (8, 11-15) for extracting starch granules from stone tools.

1. Tools were sonicated for 5 min in de-ionized, distilled water.

2. Tools were rinsed off into the beaker containing the sonicated water and residue, and placed under a bio-cabinet to dry.

3. The beaker containing the residue was covered and allowed to settle for 12 h.

4. Most of the water was decanted from the beaker, leaving »2 cm of water and residue that had settled at the bottom.

5. The remaining water and residue were poured into 15 ml test tubes and centrifuged for 15 min at 2500 rpm to concentrate the residue. The supernatant (water) was decanted and discarded.

6. Centrifuge cycles continued until all residue from one tool was concentrated in one tube.

7. A solution of Cesium chloride (CsCl) prepared to 1.79 g/cm³ density was added to the residue, and the tubes were gently agitated to mix the residue and the CsCl.

8. Samples were centrifuged for 5 min at 2500 rpm.

9. The top 1.5 cm of the supernatant was aspirated using a Pasteur pipette and deposited into a new sterile test tube. The original sample test tubes were covered and set aside for later re-extraction if desired.

10. Distilled water was added to the test tube with the aspirated sample, up to the 15 ml line to dilute the CsCl.

11. The sample was centrifuged for 15 min at 2500 rpm. One cm of supernatant was removed and discarded.

12. Water was added again to the 15 ml line to further dilute the CsCl. The sample was mixed and then centrifuged for 15 min at 2500 rpm.

13. Two centimeters of supernatant was removed, and step 12 was repeated.

14. Two more rinsing cycles were completed, each time aspirating an additional 2 cm of supernatant, until the final cycle, when supernatant was removed until only 0.7 cm of liquid remained in the bottom of the test tube, containing extracted material.

15. This extract was mounted on glass microscope slides using a simple water mount, and covered with a glass coverslip. The slides were viewed under an Olympus BX41 compound transmitted light microscope equipped with cross-polarizing filter and a digital camera.

1. Haslam, M. (2004) J Archaeol Sci 31, 1715-1735.

2. McNair, JB. (1930) The Differential Analysis of Starches.

3. Reichert, ET. (1913) The Differentiation and Specifity of Starches in Relation to Genera, Species, etc. (Carnegie Institution of Washington, Washington, D.C.).

4. Atchison, J & Fullagar, R. (1998) in A Closer Look: Recent Australian Studies of Stone Tools, ed. Fullagar, R (Archaeological Computing Laboratory, University of Sydney, Sydney), pp. 109-125.

5. Barton, H, Torrence, R & Fullagar, R. (1998) J Archaeol Sci 25, 1231-1238.

6. Fullagar, R, Loy, TH & Cox, S. (1998) in A Closer Look: Recent Australian Studies of Stone Tools, ed. Fullagar, R (Archaeological Computing Laboratory, University of Sydney, Sydney), pp. 49-60.

7. Loy, TH, Spriggs, M & Wickler, S. (1992) Antiquity 66, 898-912.

8. Pearsall, DM, Chandler-Ezell, K & Zeidler, JA. (2004) J Archaeol Sci 31, 423-442.

9. Perry, L. (2005) Lat Am Antiq 16, 409-426.

10. Piperno, DR & Holst, I. (1998) J Archaeol Sci 25, 765-776.

11. Piperno, DR, Ranere, AJ, Holst, I & Hansell, PK. (2000) Nature 407, 894-897.

12. Loy, TH. (1994) in Tropical Archaeobotany: Applications and New Developments, ed. Hather, JG (Routledge, London), pp. 86-114.

13. Therin, M, Fullagar, R & Torrence, R. (1999) in The Prehistory of Food: Appetites for Change, eds. Gosden, C & Hather, JG (Routledge, New York), pp. 438-462.

14. Fullagar, R. (2006) in Ancient Starch Research, eds. Torrence, R & Barton, H (Left Coast Press, Walnut Creek, CA), pp. 177-204.

15. Perry, L. (2004) J Archaeol Sci 31, 1069-1081.

16. Bronk Ramsey, C. (2006) OxCal 4.0 (Oxford Radiocarbon Accelerator Unit, Oxford, UK).

17. Piperno, DR. (2006) in Documenting Domestication, eds. Zeder, M, Bradley, DG, Emshwiller, E & Smith, BD (University of California Press, Berkley, CA), pp. 46-67.