Stable Isotope Analysis and Ancient Diets: They Are What They Ate R.H. Tykot, Department of Anthropology, University of South Florida. Email: [email protected]

72nd Annual Meeting, Austin, Texas Introduction Stable Isotope Studies of Human Diets Applications – Latin America Applications – Old World April 26, 2007 Stable carbon, nitrogen, and oxygen isotope studies have become widely Isotope analyses to reconstruct human dietary patterns have now been going on for over thirty years (Vogel In the last fifteen years I have worked on a large number of In Europe, it was only in the Iron Age that a C4 plant, millet, was applied to studies of ancient subsistence patterns, providing complementary & van der Merwe 1977; see Tykot 2006). Laboratory-based and other isotope studies have specifically isotope projects, on all continents. Illustrated here are just some introduced from Africa, so studies on earlier periods have focused

information to faunal, paleobotanical and other dietary indicators (Figure 1). determined that bone collagen (the organic portion, consisting of many different amino acids) is produced examples of studies done in , , , Peru, Chile, entirely on the importance of seafood and other aquatic resources. In Magdalenska Gora Using small samples of bone, teeth, and/or hair for this inexpensive analysis, the mainly from dietary protein, while bone apatite (the mineral portion, consisting of calcium Argentina, Florida, Italy, and China. Most of these have been east Asia, another variety of millet was locally domesticated by the Arene Candide

15 study of many individuals has often revealed dietary differences based on age, hydroxyphosphate/carbonate) represents the whole diet (Ambrose & Norr 1993; Tieszen and Fagre 1993). published or are in press. beginning of the Neolithic. Catignano La Selvicciola Coppa Nevigata ANAS Figure 12. A number Ripa Tetta M. Saraceno Isola Sacra Massieria Candelaro Trinitapoli Rendina 12 sex, status, and other sociocultural factors; local ecology; seasonal variation; and Turnover rates for bone are slow, so that the isotope values obtained represent at least the last several years of Toppo Madonna di Grottole (LC) Daguzzo Malerba II Sant’Abbondio Tzamala Barton Ramie (EC) of projects done in Barton Ramie (LC) Neanderthals Menorca 9 (Pre-C) 18 Britain (Upper Paleolithic) chronological change. Presented here is an overview of sampling and analysis an individual’s life. Tooth enamel and dentin, while composed of the same material as bone apatite and (Pre-C) Theopetra Lamanai (EC) the Mediterranean Oronsay (Hebrides) Formentera Lamanai (LC/TC) Tharrounia Ya x una Grotta dell’Uzzo Lamanai (Post-C) Portugal (Mesolithic) 6 Lamanai (H) 15 Mycenae

methods; examples of studies on maize in the New World, millet in the Old collagen, respectively, do not turn over, so their isotope values reflect the specific time period of their delta 15N (per mil) b 13 Portugal (Neolithic) (EC) also suggest that Kephala Preclassic * C = -9.5 ± 1.2‰ Franchthi Early Classic 13C = -6.8 ± 1.2‰ Pacbitun (LC/TC) Britain (Iron Age) * (LC/TC) Alepotrypa Kouveleiki Late/Terminal Classic 13C = -6.5 ± 1.0‰ 3 * Britain (Late Roman) Brochtorff Circle World, and the importance of seafood; and suggestions for future research. formation regardless of the age at death (e.g. Wright 1999). Isotope studies on enamel have produced reliable Santa Rita Postclassic *13C = -6.4 ± 1.8‰ 12 seafood was never a Gerani Cuello Historic *13C = -5.6 ± 0.7‰ Armenoi La Milpa Lamanai 0 dietary staple, even data going back millions of years (e.g. Lee-Thorp 2000). Hair and teeth both grow sequentially, so that Baking Pot Mojo Cay Barton Ramie -12 -9 -6 -3 9 Sites in peninsular Italy and Sardinia by Tykot and colleagues delta 13C (apatite) (per mil) Pacbitun on islands. microsampling may be used to address short term isotopic variation (e.g. O’Connell & Hedges 1999; Balasse a 15 Delta 15N (per mil) Aguate ca Archaeological Evidence for Subsistence 6 Portugal data from Lubell et al. 1994; other data from 12 2002). Baking Pot (LC) publications by Richards Copan Barton Ramie (EC) 15 Faunal remains Barton Ramie (LC) 3 La Blanca Cahal Pech (Pre-C) Overall, stable isotope analysis is able to quantify the proportions of C3 and C4 plants (e.g. maize) and their 9 Cahal Pech (LC) Cuello (Pre-C) Macrobotanical remains Lamanai (Pre-C) -24 -20 -16 -12 Lamanai (EC) -22 -18 -14 -10 c Lamanai (LC/TC) Coprolites or stomach contents consumers in human diets, compare terrestrial vs. freshwater vs. marine resources, and reveal whether there 6 Lamanai (Post-C) delta 15N (per mil) Lamanai (H) Delta 13C (per mil) 12 Preclassic d13C = -12.6 ± 1.2 Pacbitun (EC) Early Classic d13C = -11.5 ± 1.3 Pacbitun (LC/TC) Armenoi (LMIII Crete) Pollen Figure 5a,b. Isotope studies in the New World La Milpa (LC/TC) were systematic differences between individuals based on geography, chronology, social status, and/or sex, in 3 Late/Terminal Classic d13C = -11.3 ± 2.3 Postclassic d13C = -9.3 ± 0.8 Figure 11. Several studies have shown that in the Atlantic, Mycenae Circle A Tykot projects in red Historic d13C = -9.6 ± 0.5 Phytoliths distinguish maize from other plant foods Mycenae Grave B both the New and Old Worlds (Figure 5, left). 0 seafood was only important in pre-neolithic times. Mycenae (LHI-III) -18 -15 -12 -9 -6 Starch grains 9 Balearics (above); in the Old World most studies focus on delta 13C (per mil) Shell middens Sardinia the importance of aquatic resources, and the use Figure 8a-c. A large number of sites have been done now in Belize and Malta Bone chemistry mil) (per 15N Delta 13.2 of millet and secondary animal products (below). Guatemala, but relatively few in Mexico and . A comparison of 6 Skeletal pathology Female results from different time periods suggests a significant jump in maize in the Data for Crete, Mycenae and Malta from Richards; Pottery residues 12.2 Male for the Balearics from Van Strydonck et al. 2002. overall diet by the Early Classic, and possibly as a protein source by the Artistic depictions 11.2 -24 -20 -16 Historical documentation Postclassic. -22 -18 10.2 Delta 13C (per mil)

Plant impressions on ceramics 15 N delta 9.2 Figure 13. Isotope data available for inland and coastal sites Evidence of land clearance, irrigation not shown: 8.2 in Italy, and on Sardinia, Malta, the Balearics, and Crete, Argentina: Buque Sur, Bajo de la Quinta Peru: Chokepukio -20.2 -19.9 -19.6 -19.3 -19 -18.7 -18.4 -18.1 Peru: Lurin mummies show little if any seafood consumption. For the inland site of Figure 1. Isotope analyses strongly complement other Figure 6. Sample selection and preparation in the Laboratory for Archaeological Science at the University of South Florida. Peru: Chongos hair delta 13 C Venezuela: Parmana Mycenae, however, the highest status individuals appear to evidence about ancient dietary practices One gram of bone is used to produce collagen samples for analysis, and less than 10 mg for apatite and tooth enamel . Cerro Mangote Socapamba Figure 14. On Sardinia, at the Bronze Age site of La Chimba have had some access to seafood. Im-11 Loma Alta La Florida Las Vegas Lake Ayauch Padru Jossu, there appears to be small but measurable Real Alto La Emerenciana PacopampaGuitarrero male-female differences in diet. CerroViru Oreja Valley Huaricoto Casma Valley Waman Wa in Isotope Basics Los Gavilanes Chavin de Huantar Applications - North America Cardal Jauja Mina Perdida Tablada de Lurin Machu Picchu 16 Isotopes are elements which vary in the number of neutrons in their nuclei Villa El SalvadorOsmore Valley Tiwanaku 16 Khonkho Wankane Pisagua Camarones Tarapaca Data in red from Pechenkina et al. 2005 (Figure 2). The multiple isotopes of carbon (12, 13, 14), nitrogen (14, 15), and Tiliviche 15 San Pedro 14 Weeden Island oxygen (16, 17, 18) affects the energy used in chemical reactions, resulting in a Huachichocana 14 Pillsbury

12 variation in isotope ratios in different materials (Tykot 2004). Except for carbon Melton Mound I Central Chile San Juan Pillsbury Mendoza b 13 Arroyo Seco Horr's Island 14, these are all stable isotopes which do not decay over time. Melton Mound I Dunwoody 10 d15N Collagen d15N Horr's Island North-Central Patagonia d15N Collagen Micro-sampling of 12 Crystal River Dunwoody

Sample Preparation and Stable Isotope Analysis prehistoric Ecuador incisor Cross Creek 8 Crystal River 11 shows 4‰ carbon range Bayshore Homes Bayshore Homes Less than one gram of bone is necessary to extract sufficient collagen (a 1% yield is Tierra del Fuego 500 Km 6 Bay Pines 10 Bay Pines protons + neutrons 500 Mi. -22 -20 -18 -16 -14 -12 -10 -8 -6 -16 -14 -12 -10 -8 -6 -4 considered reliable) for stable isotope analysis, and may generally come from any part of the Tykot projects in red d13C Collagen d13 C Bone Apatite skeleton. Much smaller samples of well-preserved tooth dentin may also be tested, while for d Figure 15. In northern China, millet appears to have Figure 16. Many studies have been done in North America, apatite and enamel analyses only several milligrams are needed. Overall, isotopic analysis is 18 decreased in importance in later periods, perhaps with mainly looking at the spread of maize from the southwest to as 4001 number of protons 3005 4002 minimally destructive to skeletal remains. 3004 16 30033001 the expansion of rice from the south. The values shown far as Canada. One of the studies done at USF compared inland, 3002 4003 5204 At the University of South Florida, sample preparation is done using well-established 1202 for Liangchengzhen are based on apatite (hence no riverine, and coastal sites in peninsular Florida, with the bone 1203 5203 47034702 4701 Ind. 12 14 4704 laboratory procedures (Figure 6, above). Bone collagen is extracted by demineralizing whole 1204 5202 Ind. 18 nitrogen isotope values). apatite values suggesting the consumption of maize by 600 AD. 1201 4004 5201 Ind. 30 bone using 2% HCl for 72 hrs, dissolving base-soluble contaminants using 0.1 M NaOH (24 hrs c 12 Ind. 40 Figure 2. Carbon atom and elemental signatures 1805 Ind. 47 1804 Ind. 51 before and after demineralization), and separating residual lipids with a mixture of methanol, delta 15N (per mil) 10 Ind. 52 Other Stable Isotope Studies 1803 chloroform and water for 24 hrs. 51065107 5101 5103 5104 12 8 5105 Carbon dioxide (CO ) in the atmosphere consists of about 98.9% CO and * (‰) = [R /R ) - 1] x 1000 5102 1802 Other isotope analyses recently applied to archaeology include specific studies on soils (e.g. Webb et al. 2004); 2 2 sample standard Collagen pseudomorphs are analyzed for carbon and nitrogen isotopes using a CHN analyzer 1801 13 14 -10 R is the ratio of the heavier to the lighter isotope 1.1% CO2 (plus radioactive CO2 about 1 x 10 %). CO2 is photosynthesized 6 movement and migration using oxygen (e.g. White et al. 2000) and strontium (e.g. Price et al. 2002) isotopes; and coupled with a Finnigan MAT Delta Plus XL stable isotope ratio mass spectrometer set up with a -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9 Hair segments from Chongos by plants and metabolized into complex molecular compounds that we Figure 3. Carbon isotope fractionation from continuous flow (Figure 7, below). Along with visual analyses and data from the sample delta 13C (per mil) (Peru) show C and N differences compound specific analysis of human bones (e.g. Howland et al. 2003) and pottery residues (e.g. Reber & Evershed categorize as carbohydrates, proteins, and lipids. Lighter isotopic compounds atmospheric to plants to consumers preparation, C:N ratios of the analyzed gases are calculated to determine the preservation of 2004). All of these isotope methods may be used to address significant questions we have about ancient societies in react faster, using less energy, resulting in changes in the 13C/12C ratio from the Figure 9a-d. In South America, a number of major isotope studies have been many parts of the world. collagen and the reliability of the isotope results. done in recent years, at least along the Andes and in Patagonia (a). atmosphere to plants. This is called isotopic fractionation, and results in Carbonate from apatite and enamel samples is also extracted using established techniques, with References Microsampling of teeth (b) from highland Ecuador, and of hair (c, d) from ¾Ambrose, S.H. & L. Norr 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In J. B. Lambert & G. different carbon isotope ratios between plants with different photosynthetic Grupe (eds.), Prehistoric Human Bone: Archaeology at the Molecular Level, pp. 1-37. New York, Springer-Verlag. the removal of organic components using bleach (24 hrs for enamel, 72 hrs for apatite), and of Peru show significant short-term dietary variation. ¾Balasse, M. 2002. Reconstructing dietary and environmental history from enamel isotopic analysis: time resolution of intra-tooth sequential sampling. International Journal of Osteoarchaeology 12: 155- pathways (C3, C4, and CAM), and this difference is passed on to their 165. non-biogenic carbonates using buffered 1 M acetic acid (24 hrs). These samples are introduced to ¾Howland, M. R., L. T. Corr, S. M. M. Young, V. Jones, S. Jim, N. J. van der Merwe, A. D. Mitchell & R. P. Evershed. 2003. Expression of the dietary isotope signal in the compound-specific C13 consumers (Figure 3, right). 25 values of pig bone lipids and amino acids. International Journal of Osteoarchaeology 13: 54-65. a second Delta Plus XL mass spectrometer using an automated Kiel III device. 16 15 ¾Lee-Thorp, J.A. 2000. Preservation of biogenic carbon isotopic signals in Plio-Pleistocene bone and tooth mineral. In S.H. Ambrose & M.A. Katzenberg (eds.), Biogeochemical Approaches to 15 14 * N(‰) Paleodietary Analysis, pp. 89-116. Plenum, New York. Stable nitrogen ( N/ N) isotopes vary between nitrogen fixing and non- ¾Lubell, D., M. Jackes, H. Schwarcz, M. Knyf & C. Meiklejohn. 1994. The Mesolithic-Neolithic transition in Portugal: isotopic and dental evidence of diet. Journal of Archaeological Science 21: 201-16. 14 green = coastal blue = valle red = precordillera 18 16 20 ¾O'Connell, T.C. & R.E.M. Hedges. 1999. Isotopic comparison of hair and bone: archaeological analyses. Journal of Archaeological Sciences 26: 661-665. fixing plants, and along with oxygen ( O/ O) isotope ratios also vary based on ¾Pechenkina, E.A., S.H. Ambrose, M. Xiaolin & R.A. Benfer. 2005. Reconstructing northern Chinese Neolithic subsistence practices by isotopic analysis. Journal of Archaeological Science 32: 1176- 12 Arcaico - Costa 1189. Arcaico - Precordillera 15 ¾Price, T.D., J.H. Burton & R.A. Bentley. 2002. The characterization of biologically available strontium isotope ratios for the study of prehistoric migration. Archaeometry 44: 117-35. climate and altitudinal effects and between terrestrial, freshwater, and marine Bato - Costa ¾Reber, E.A. & R.P. Evershed. 2004. How did Mississippians prepare maize? The application of compound-specific carbon isotope analysis to absorbed pottery residues from several Mississippi Valley 10 Bato - Valle sites. Archaeometry 46: 19-33. ecosystems. When plant and animal foods are consumed, the metabolic PAT - Precordillera ¾Tieszen, L.L. & T. Fagre 1993. Effect of diet quality and composition on the isotopic composition of respiratory CO2, bone collagen, bioapatite, and soft tissues. In J..B. Lambert & G. Grupe (eds.), 10 N. Tehuelches Prehistoric Human Bone: Archaeology at the Molecular Level, pp. 121-155. New York, Springer-Verlag. 8 PAT - Valle S. Tehu elches processes involved fractionate the isotope ratios, increasing the proportion of the Llolleo - Costa Selk'nam ¾Tykot, R.H. 2004. Stable isotopes and diet: you are what you eat. In Physics Methods in Archaeometry (M. Martini, M. Milazzo & M. Piacentini eds.), 433-444. Bologna, Italy. Llolleo - Valle Haush ¾Tykot, R.H. 2006. Isotope Analyses and the Histories of Maize. In J.E. Staller, R.H. Tykot & B.F. Benz (eds.), Histories of Maize: Multidisciplinary Approaches to the Prehistory, Linguistics, delta 15N collagen 15N delta 6 Inland sites: squares Yaman a Biogeography, Domestication, and Evolution of Maize, 131-142. Academic Press (Elsevier). heavier carbon (and nitrogen) isotopes remaining in the body tissues. When Aconcagua - Coastal 5 Alakaluf Aconcagua - Valle Coastal sites: circles ¾Van Strydonck, M., M. Boudini & A. Ervynck. 2002. Stable isotopes and diet: Animal and human bone collagen from prehistoric sites on Mallorca, Menorca, and Formentera (Balearic Islands, Spain). In W.H. Waldren & J.A. Ensenyat (eds.), World Islands in Prehistory, pp. 189-197. BAR International Series 1095. Oxford: Archaeopress. 4 Inca - Valle carbon and nitrogen isotopes are graphed together, several different food groups ab¾Vogel, J.C. & N.J. van der Merwe. 1977. Isotopic evidence for early maize cultivation in New York State. American Antiquity 42 (1977) 238-242. 0 ¾Webb, E.A., H.P. Schwarcz & P.F. Healy. 2004. Detection of ancient maize agriculture in the Maya lowlands using the stable carbon isotope compositions of soil organic matter: Evidence from Caracol, 2 may be differentiated (Figure 4, right). -25 -15 -5 Belize. Journal of Archaeological Science 31: 1039-1052. -22 -20 -18 -16 -14 -12 -10 -8 -20 -10 ¾Wright, L.E. 1999. Correspondence between stable carbon, oxygen and nitrogen isotopes in human tooth enamel and dentine: infant diets at Kaminaljuyu. Journal of Archaeological Science 26: 1159-70. 13 Carbon, oxygen, and nitrogen isotope ratios are reported using the delta (δ) delta 13C collagen * C (‰) ¾White, C.D., M.W. Spence, F.J. Longstaffe & K.R. Law. 2000. Testing the nature of Teotihuacán imperialism at Kaminaljuyú using phosphate oxygen-isotope ratios. Journal of Anthropological Research 56: 535-58. notation, in parts per mil (‰) relative to the PDB (C, O) and AIR (N) Figure 4. Graph showing the carbon and nitrogen isotope Figure 7. Marine Sciences Laboratory at USF St. Petersburg. From left: one mass spectrometer, with CHN Figure 10. A regional study in central Chile (a) shows geographic and chronological differences, with Acknowledgments standards. Plant and animal compounds simply have less of the heavier isotopes Financial support for my isotope studies has come directly, or through my colleagues, from the National Science Foundation, the Wenner-Gren Foundation, the Florida Anthropological Society, numerous values associated with several food groups. Using more analyzer for organic samples; second mass spectrometer, with Kiel device for inorganic carbonate samples. maize a staple in the PAT, Llolleo and Aconcagua periods. In prehistoric Patagonia (b), the relative foreign universities and agencies, and many internal awards from the University of South Florida. The many students who have worked in my lab include: Lori Adornato, Jonathan Auth, Robert Bowers, Leya Collins, Beau DeBoer, Annette Doying, Jennifer Dukes, Cristina Echazabal, Nicole Falk, Jack Fenner, Scott Grammer, Jason Hicks, Jennifer Kelly, Steve Kosiba, Kathryn Lafrenz, Luca Lai, Rheta than atmospheric CO2, resulting in relative isotope ratios that are negative. specific values for particular localities is recommended. Both systems allow the input of several dozen samples and standard reference materials importance of seafood and guanaco corroborates some ethnohistoric accounts. Lanehart, Eugenia Mansell, Anna Metroka, Jose Moreno-Cortes, Michael Pateman, Stephanie Roberts, Nicole Shelnut, Charles Summers, Dolores Tedesco, Michelle Whitaker, Jason Wilson.