Journal of Archaeological Science 36 (2009) 1312–1318 Contents lists available at ScienceDirect Journal of Archaeological Science journal homepage: http://www.elsevier.com/locate/jas Stable isotope evidence of human diet at the Nukdo shell midden site, South Korea Kyungcheol Choy a, *, Michael P. Richards a, b a b Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, Leipzig 04103, Germany Department of Archaeology, University of Durham, South Road, Durham DH1 3LE, United Kingdom a r t i c l e i n f o a b s t r a c t Article history: Received 26 August 2008 Received in revised form 19 December 2008 Accepted 6 January 2009 Carbon and nitrogen stable isotope analysis was undertaken on bone collagen extracted from archaeological human (n ¼ 48) and animal (n ¼ 45) skeletons from the Nukdo site, Location I C, South Korea. This shell midden and grave site is dated from the late Mumun (550–300 BC) to early Iron Age (300 BC-1 AD) periods. The herbivorous mammals fell within the range of C3 consumers, with average values of d13C ¼ 21.0  0.5& and d15N ¼ 3.6  0.5& for deer (n ¼ 16) and d13C ¼ 20.6  0.5& and d15N ¼ 4.5  2.0& for wild boar (n ¼ 17). Humans from this site averaged d13C ¼ 18.3  0.4& and d15N ¼ 11.2  0.7& for adults (n ¼ 15) and d13C ¼ 18.7  0.7& and d15N ¼ 12.5  1.1& for juveniles (n ¼ 33). These d13C values indicate that there was no significant input of C4 plants in the human diets and this may be associated with the spread of rice agriculture in the Mumun period. Human bone collagen d13C and d15N values indicate that there was some consumption of marine foods, although the main protein sources were from terrestrial foods. The isotope data demonstrate that the humans at Nukdo had mixed diets that included marine and terrestrial protein, including C3 plants such as rice. Finally, the isotope results from the juveniles indicate that weaning occurred before the age of 1.5 years in this period. Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved. Keywords: Carbon Nitrogen Stable isotopes Palaeodiet Weaning Mumun Nukdo South Korea 1. Introduction The Mumun period of Korea (1500–300 BC) is characterized by intensive agriculture and a complex society, with megalithic burials, large settlements, and the undecorated Mumun pottery (Bale, 2001; Rhee and Choi, 1992). After the spread of domesticated plants from central China, the transition to agriculture in Korea occurs in the middle Jeulmun period (3500–2000 BC) with the introduction of foxtail millet and rice (Crawford and Lee, 2003; Crawford and Shen, 1998). At the beginning of the Mumun period, plant cultivation is intensified and spreads to all regions, eventually diffusing to the Japanese Archipelago (Crawford and Lee, 2003). However, Mumun sites also have many shell midden sites similar to sites dating to the previous Jeulmun culture (8000–1500 BC). This means that there is still a lot of exploitation of marine and terrestrial foods after the spread of agriculture. The nature and significance of husbandry and agriculture on prehistoric diets in the Korean peninsula are poorly understood. Therefore, investigating food consumption in prehistoric Korea will help to illuminate prehistoric subsistence activities and the spread of agriculture in East Asia in general. Stable carbon and nitrogen isotope analysis is a well-established method for reconstructing past human diets (Lee-Thorp, 2008; * Corresponding author. Tel.: +49 03413550 771; fax: þ49 03413550 399. E-mail address: choy@eva.mpg.de (K. Choy). Sealy, 2001) and provides us with direct, long-term evidence of the sources of dietary protein. Although there is an increasing application of stable isotope techniques to archaeological research, very little has been applied to Korean archaeological bone remains. We present here the application of stable isotope analyses to human and animal bones excavated from the Nukdo shell midden site in South Korea. We obtained and analyzed human bones from many different burial types including jar coffins, stone-cist coffins, and wood-cist coffins. Faunal remains from the site were examined in order to obtain an isotopic baseline. In this paper, we examine the significance of terrestrial mammals and seafood in the human diet between the late Mumun and early Iron Age period, associated with the spread of agriculture. Additionally, the analysis of juveniles from jar coffins was conducted to determine the age of weaning. 2. The Nukdo site The Nukdo site (ND) is located on an island in the southern Korean peninsula (Fig. 1). The first and second excavations of this site were performed by the Busan National University Museum during two field seasons in 1985 and 1986, and a third excavation was undertaken in 1998–2001 (Busan University Museum, 1989, 2004). This is one of the largest archaeological sites in Korea, and exposed graves as well as the remains of shell middens and 0305-4403/$ – see front matter Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jas.2009.01.004 K. Choy, M.P. Richards / Journal of Archaeological Science 36 (2009) 1312–1318 1313 Fig. 1. The location map of the Nukdo site and the other Japanese sites mentioned in this study. The sites are as follows: 1: Nukdo site, 2: Ohtomo site, 3: Ikiriki site, 4: Awazu site. settlements. Based on archaeological records and cultural features, this site was occupied from the Mumun period up to the ProtoThree Kingdoms period (4th century BC–1st century AD). The recovery of Chinese coins and Yayoi pottery in the third excavation shows that the Nukdo site was the main channel connecting China and Japan during the Bronze Age (Lee, 2004; Suzuki et al., 2008). The excavations have yielded over 74 human burials dating to the early Iron Age period (300 BC–1 AD). Most of these graves were located within the location I C in the southeastern part of the island. Among the 74 graves, there were 33 jar coffins for infants and children, and 39 wood and 2 stone-cist coffins for adults. Human remains were relatively well preserved under the shell mounds and the preservation of infant bones within jar coffins was good. Most of the burials had associated grave goods such as pottery, iron tools, and grindstones. Physical anthropological assessments of the human skeletons were undertaken by the Busan University Medical College and the Kagoshima University Dental School (Kim et al., 1988, 1990). In addition to the human skeletons this excavation also revealed many animal remains. According to the faunal analysis, deer (Cervus nippon hortulorum) and wild boar (Sus scrofa) were the dominant species among the terrestrial animals and deer forms about 90% of the fauna (Kaneko et al., 1990; Kaneko and Seo, 2004). In addition, there were a few specimens of bear, bird and rat. Some middle-sized dog bones were also excavated. The only marine carnivores were two sea lion bones which probably mean that there was less marine animal hunting in this period than in the previous Jeulmun period (Kaneko, 2004). Additionally, the burials found under the shell middens contained a variety of shellfish, mostly oyster (Crassostrea gigas) and gastropods (Lunella coronata coreensis). A few fish bones were excavated and the main fish species present were bream (Sparidae). Only the jar coffin burial ND5 contained many animal bones and it is assumed that animals in the jar coffin were added intentionally as grave goods (Kaneko and Seo, 2004). The main species represented in the coffin were deer and wild boar like the rest of the site, but marine species (bream and sea lion) as well as terrestrial mammals (bear and pheasant) were also present. 3. Stable isotope analyses for determining palaeodiet Stable isotope analysis for dietary reconstruction is based on the principle that the isotope values of consumed food are reflected in the isotope composition of the body. C and N isotope analysis is one of the best-established techniques to detect past human diets (LeeThorp, 2008; Sealy, 2001). Carbon stable isotopes are suitable for distinguishing food consumption of terrestrial vs. marine animals, and C3 vs. C4 plants (Schwarcz and Schoeninger, 1991). For instance, rice (Oryza sativa) is a C3 plant and the d13C value of rice is 26&, but foxtail millet (Setaria italica) is a C4 plant and the d13C signature of millet is about 11& (Hu et al., 2006; Pechenkina et al., 2005). This means that we can differentiate between the consumption of rice and millet, both of which are important crops in this region associated with the development of agriculture. Nitrogen stable isotopes reflect the trophic level and can be used to reconstruct food webs (Schoeninger and DeNiro, 1984). Furthermore, it can 1314 K. Choy, M.P. Richards / Journal of Archaeological Science 36 (2009) 1312–1318 enable us to investigate the influence of marine protein in the diet (Chisholm et al., 1982; Schoeninger et al., 1983). Humans that obtain their protein from marine food typically have d13C values close to 12& and d15N values between 12& and 22&, while humans that consume only C 3 based terrestrial protein sources have d13C values of about 20& and d15N values ranging from 5& to 12&. Humans that have a mixture of marine and C3 based terrestrial protein would have isotope values somewhere between those end points (Richards et al., 2006; Schoeninger and Moore, 1992). The d15N values of human bone collagen are enriched by approximately 3–5& compared to the d15N ratio of the protein that the human has consumed (Hedges and Reynard, 2007; Bocherens and Drucker, 2003). On the basis of the difference in nitrogen isotope values between mothers and children, nitrogen isotope values also provide an opportunity to detect the timing of weaning (Fuller et al., 2006; Katzenberg and Pfeiffer, 1995; Schurr, 1997). In archaeological research, carbon and nitrogen stable isotope ratios are obtained from bone collagen, since bone collagen is often well preserved in bone after burial. However, collagen turnover rates vary between different types of bone and the isotope composition of bone collagen reflects an average of the dietary protein consumed over the last 10–30 years of life (Ambrose, 1993; Schwarcz and Schoeninger, 1991). Therefore, bone collagen provides a long-term record of food consumption, primarily protein, in individuals. 4. Materials and methods Bone samples from 48 human and 45 animal skeletons were selected from the Nukdo collections in the Busan National University, South Korea. Preparation and isotope measurement of bone samples were undertaken in the isotope analysis laboratory, Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology. To obtain bone collagen from each sample, we followed the protocol outlined in Richards and Hedges (1999) with the addition of an ultrafiltration step (Brown et al., 1988). The Nukdo samples were prepared as follows. For each test, 500 mg of bone powder was put into a 0.5 M HCl solution at 5  C for two weeks, and then gelatinized at 70  C in a pH 3 solution (Richards and Hedges, 1998). After removing insoluble residues with a 5–8 mm EzeeÒ mesh, the remaining solution was concentrated in UltrafreeÒ-4 Centrifugal Filter Units fitted with Biomax-30 membranes (Millipore) (Brown et al., 1988). The supernatant, purified collagen (>30 kDa), was freeze-dried for 48 h. The isotope values were measured using an automated carbon and nitrogen Table 1 Faunal stable isotope values from the Nukdo site, Korea (ND-L ¼ Nukdo Level Number; ND-B ¼ Nukdo B Location; UI ¼ Unidentified). S-EVA Location Species Element Collagen weight (mg) d13C& d15N& %C %N C:N 1897 1898 1899 1900 1901 1902 1904 1905 1906 1907 1917 1922 1928 1929 1930 1931 1903 1908 1909 1910 1912 1913 1914 1915 1916 1918 1919 1920 1921 1923 1924 1925 1926 1911 3928 3929 3930 3932 3934 3935 3938 3937 3936 3933 3931 ND-L3 ND-L3 ND-L4 ND-L29–4 ND-L50 ND-L4 ND-L4 ND-L5 ND-L6 ND-L3 ND-L65 ND-L6 ND-L8 ND-L6 ND-L25 ND-L22 ND-L37 ND-L6 ND-L5 ND-L5 ND-L35 ND-L15 ND-L54 ND-L68 ND-L3 ND-L66 ND-L3 ND-L4 ND-L5 ND-L5 ND-L22 ND-L7 ND-L34 ND-L68 ND-B Na-1b ND-B Na 137 ND-B Da 50 ND-B Na Id-1 ND-I a ND-2 B5 ND-I a ND-I a ND-B Na 3b 2 ND-B Na 4a-W2 ND-B Na 3a 4a Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Cervus nippon hortulorum (Deer) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Sus scrofa (Wild boar) Canidae (Dog) Canidae (Dog) Canidae (Dog) Bovidae (Cattle) Ursidae (Bear) Rattus (Rat) Bird Bird Wild duck Mugilidae (Sea mullet) Sparidae (Porgy) Zalophus japonicus (Sea lion) Tarsal Humerus Humerus Humerus Femur Femur Humerus Humerus Humerus Humerus Tarsal Humerus Tarsal Tarsal Tarsal Tarsal Ulna UI Mandible Mandible Humerus Radius Scapula Ulna Humerus Ulna UI Radius Scapula Humerus Metatarsal Metatarsal Metatarsal UI UI UI UI UI UI UI UI UI UI UI UI 11.1 2.5 2.1 3.3 8.0 8.2 3.5 2.9 6.5 7.0 14.3 9.7 13.9 6.9 1.9 8.6 5.6 8.8 5.2 7.0 10.8 4.2 8.6 8.4 2.0 7.9 7.5 12.5 13.1 5.8 12.7 3.8 8.3 4.9 13.6 18.8 11.4 9.6 10.3 5.8 7.6 18.4 7.2 10.2 12.8 21.0 20.9 21.8 21.8 20.5 20.4 20.8 20.4 20.8 20.9 22.1 21.0 21.1 20.7 21.3 20.9 21.1 20.5 20.9 20.9 19.9 20.0 20.1 20.9 21.1 20.3 20.4 21.1 21.2 19.8 21.1 21.5 20.5 18.1 18.6 15.6 16.0 20.1 17.3 20.3 10.0 20.8 8.7 10.8 12.5 3.9 4.7 4.3 3.0 3.8 3.5 3.0 3.1 3.4 3.5 3.3 4.0 3.3 3.9 3.9 3.5 3.0 4.5 3.3 3.1 9.7 5.0 4.6 2.7 3.4 6.9 4.9 3.4 3.5 8.5 3.5 3.2 3.8 11.7 10.1 12.5 6.8 3.5 8.7 5.5 12.7 8.1 9.6 13.8 16.4 40.1 44.1 23.1 38.9 42.3 41.9 32.3 44.6 39.1 43.8 44.4 44.4 39.2 43.4 36.5 42.3 39.9 42.4 36.7 42.2 42.8 34.1 40.3 40.3 23.0 41.9 42.2 44.5 44.7 39.1 42.9 36.2 42.0 40.6 43.7 43.6 43.4 43.2 44.2 38.2 41.4 44.9 41.2 43.6 41.9 14.7 15.6 8.3 14.0 15.2 15.1 11.6 15.8 14.1 15.7 16.4 16.1 14.6 16.0 13.2 15.6 14.5 15.1 13.2 15.3 16.3 13.1 15.4 15.5 8.9 15.3 15.3 16.3 16.3 14.2 15.6 13.1 15.4 15.1 16.2 16.0 15.8 15.8 16.2 13.6 14.7 16.0 14.4 16.1 15.6 3.2 3.3 3.2 3.3 3.2 3.2 3.2 3.3 3.2 3.3 3.2 3.2 3.1 3.2 3.2 3.2 3.2 3.3 3.3 3.2 3.1 3.1 3.1 3.1 3.0 3.2 3.2 3.2 3.2 3.2 3.2 3.2 3.2 3.1 3.2 3.2 3.2 3.2 3.2 3.3 3.3 3.3 3.4 3.2 3.1 1315 K. Choy, M.P. Richards / Journal of Archaeological Science 36 (2009) 1312–1318 elemental analyzer coupled to a continuous-flow isotope-ratiomonitoring mass spectrometer. The elemental analyzer provided the carbon and nitrogen contents and the C/N ratio was used for checking the purity of extracted collagen. The d13C and d15N values were measured relative to the Pee Dee Belemnite (vPDB) carbonate and atmospheric nitrogen (AIR) standards, respectively. Conventional replicate measurement errors of standards were 0.1& for d13C and 0.2& for d15N. To confirm that we were measuring intact collagen, we used samples with a C/N ratio between 2.9 and 3.6 (Ambrose, 1993; DeNiro, 1985). 5. Results The stable isotope results for the Nukdo humans and animals are presented in Tables 1, 2 and Fig. 2. 5.1. Faunal isotope data The d13C and d15N ratios for terrestrial herbivores are generally very consistent. Stable isotope values of deer range mainly from 22.1& to 20.4& for carbon and from 3.0& to 4.7& for nitrogen. Isotope values from wild boar range from 21.5& to 19.7& for carbon and from 2.7& to 9.7& for nitrogen. The herbivorous mammals fell within the range of C3 consumers, averaging to d13C ¼ 21.0  0.5& and d15N ¼ 3.6  0.5& for deer (n ¼ 16), and d13C ¼ 20.6  0.5& and d15N ¼ 4.5  2.0& for wild boar (n ¼ 17). These isotope values indicate that the food source animals in the area did not consume C4 plants and instead the deer and wild boar samples have values indicating mainly C3 plant diets. However, the wild boars have a very wide range of d13C and d15N values, indicating that some of them consumed a more diverse range of food, apparently including marine protein. The d13C and d15N values of three wild boars (S-EVA1912,1918,1923) are similar to those values of the single rat we measured. This may mean that the three wild boars might be domesticated pigs or they consumed human food or waste. Furthermore, one of the cattle has higher carbon and nitrogen values than those of the deer and wild boar, which again might indicate a human influence on its diet or perhaps it was grazing on plants near the shore. Table 2 Human stable isotope values from the Nukdo site, Korea (ND ¼ Nukdo; UI ¼ Unidentified). S-EVA Location Coffin type Sex Age Element Collagen weight (mg) d13C& d15N& %C %N C:N 1938 1959 1960 1970 1981 1968 1978 1941 1945 1961 1932 1967 1971 1972 1985 1969 1973 1934 1943 1965 1946 1947 1950 1979 1951 1975 1977 1980 1984 1937 1953 1974 1958 1935 1936 1939 1942 1944 1952 1954 1955 1956 1963 1964 1966 1976 1982 1983 ND 11 ND 39 ND 40 ND 61 ND 72 ND 56 ND 69 ND 14 ND 23 ND 41 ND 1 ND 54 ND 62 ND 63 ND 74 ND 60 ND 64 ND 3 ND 17 ND 50 ND 25 ND 26 ND 29 ND 70 ND 30 ND 66 ND 68 ND 71 ND 73 ND 9 ND 32 ND 65 ND 37 ND 7 ND 8 ND 12 ND 16 ND 22 ND 31 ND 33 ND 34 ND 35 ND 46 ND 47 ND 51 ND 67 ND IA-8 ND IA-14 Jar Jar Jar Jar Stone-cist Jar Jar Jar Jar Wood-cist Jar Jar Jar Wood-cist Wood-cist Wood-cist Wood-cist Jar Wood-cist Jar Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Jar Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist Wood-cist UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI UI F F F F M F F F F M F M M M F UI UI 1M 1M 1M 1M 1M 2.5 M 2.5 M 6M 6M 6M 7.5 M 7.5 M 7.5 M 7.5 M 7.5 M 10.5 M 1Y 1.5 Y 1.5 Y 1.7 Y 3Y 3Y 3Y 3Y 3.5 Y 3.5 Y 3.5 Y 3.5 Y 3.5 Y 4.5 Y 9Y 13 Y 18 Y Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult UI UI Skull Limb bone UI skull Rib bone Rib bone Limb bone Limb bone Rib bone Rib bone Skull Rib bone Skull Rib bone Rib bone Limb bone Limb bone Limb bone Femur Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Vertebra Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Rib bone Limb bone Rib bone Femur Femur 3.7 2.2 1.2 10.1 7.3 5.3 4.6 0.7 2.6 2.5 1.5 7.5 6.2 4.9 4.1 3.8 8.7 1.0 3.3 2.5 5.5 3.3 3.7 2.1 5.5 5.5 4.7 2.7 4.5 7.8 6.7 3.6 6.5 1.7 1.7 2.8 3.0 4.7 14.7 8.4 3.1 4.7 4.3 5.8 5.7 12.2 5.1 7.3 18.4 20.3 20.1 19.0 19.2 17.4 18.6 18.2 18.2 19.7 18.3 18.3 17.9 18.5 18.8 18.1 18.4 18.8 19.8 18.6 17.7 18.4 18.0 19.6 18.0 18.0 19.5 19.5 19.5 18.1 18.9 18.3 17.9 18.8 18.0 17.9 18.5 18.8 17.4 18.3 18.5 18.5 18.2 18.4 18.3 19.0 18.5 19.2 11.9 11.3 11.8 12.5 11.3 12.3 12.3 13.7 12.6 12.6 13.4 13.3 13.7 14.3 14.0 14.4 14.6 12.6 12.2 12.6 12.6 12.2 11.8 11.3 13.8 12.6 11.1 13.2 10.7 10.6 10.8 11.7 10.9 10.7 10.8 13.4 10.2 11.7 11.3 10.7 11.0 10.7 10.8 11.9 11.2 10.8 11.3 10.5 42.4 37.5 37.9 40.9 42.9 39.2 33.8 37.0 39.9 38.4 35.8 41.9 42.0 38.9 33.8 38.4 39.4 32.8 39.4 39.2 43.5 44.5 42.7 35.4 41.7 40.3 38.8 36.0 37.6 44.2 42.2 38.5 38.7 38.8 42.3 42.8 41.7 39.4 42.7 41.7 39.3 34.1 41.4 41.7 41.3 42.8 37.2 38.2 15.1 12.6 12.5 14.3 15.0 13.7 11.5 12.7 13.8 13.3 12.3 14.9 15.0 13.7 11.9 13.7 13.9 11.3 13.4 13.7 15.4 15.1 15.1 12.1 14.7 14.2 13.7 12.2 13.1 16.0 15.1 13.6 13.4 13.9 15.4 15.2 14.5 14.0 15.5 15.0 13.8 12.0 14.4 14.7 14.7 15.5 13.3 13.4 3.3 3.5 3.5 3.3 3.3 3.3 3.4 3.4 3.4 3.4 3.4 3.3 3.3 3.3 3.3 3.3 3.3 3.4 3.4 3.3 3.3 3.4 3.3 3.4 3.3 3.3 3.3 3.5 3.3 3.2 3.3 3.3 3.4 3.3 3.2 3.3 3.3 3.3 3.2 3.2 3.3 3.3 3.4 3.3 3.3 3.2 3.3 3.3 1316 K. Choy, M.P. Richards / Journal of Archaeological Science 36 (2009) 1312–1318 5.2. Human isotope data Stable isotope results of human adult collagen extracted from this site range from d13C ¼ 22.3& to 17.4& and from d15N ¼ 11.3& to 12.3&, averaging d13C ¼ 18.3  0.4& and d15N ¼ 11.2  0.7&. The d13C and d15N results for identifiable adult males (n ¼ 5) and females (n ¼ 8) over the age of 20 years are plotted in Fig. 3. The range of d13C results indicates that the protein sources for this population were mostly terrestrial in origin, and that there was no detectable input of C4 plant foods in the diet. There is no difference between the mean d13C values of males and females. However, nitrogen values are significantly different between males and females (Fig. 3). Average d15N results of males are approximately 1& higher than those of females. The age of the infants and children studied here ranges from 1 month to 18 years. The carbon and nitrogen isotope results for all individuals analyzed from the Nukdo site are presented in Table 2. Carbon isotope values in infants and children (n ¼ 33) range from d13C ¼ 19.2& to 17.4& and from d15N ¼ 10.2& to 13.4&, with a mean of d13C ¼ 18.7  0.7& and d15N ¼ 12.5  1.1&. The d13C and d15N ratios values for infants and children are plotted in Figs. 4, 5 and adult male and female means are indicated on the graphs. 14 13 δ 15NAIR The three dog bones have quite different d13C and d15N values. Two canid bones have similar isotope values to those of most of the humans, which is a common pattern observed in isotope studies of domesticated dogs (Richards et al., 2003; Cannon et al., 1999). The third dog sample has a more positive d13C value than the others, and this is likely due to eating meat of an animal that consumed marine food. Alternatively, it is possible that this specimen is a fox, or even a raccoon. Carbon and nitrogen isotope results from bear are similar to the wild boar values. Isotope data obtained from the two birds are highly variable, reflecting different ecological adaptations or environments. In marine mammals, d13C and d15N values from sea lion (Zalophus japonicus), porgy (Sparidae), and sea mullet (Mugilidae) are as expected, with average marine d13C and d15N ratios. Isotope values from the sea lion are d13C ¼ 12.5& and d15N ¼ 16.4&. Isotope results of the porgy are d13C ¼ 10.8& and d15N ¼ 13.8&, and those of the sea mullet are d13C ¼ 8.7& and d15N ¼ 9.6&. The sea lion d15N values are approximately 3& higher than the porgy and 6& higher than the mullet, indicating that these fish were the main food sources for the sea lion. 12 11 Male Female Male Mean Female Mean 10 9 -19.2 -18.8 -18.4 -18.0 -17.6 -17.2 δ 13CvPDB Fig. 3. All adult (>20 years) male and female collagen carbon and nitrogen isotope values. 6. Discussion The carbon and nitrogen isotope values from the animal bone samples can provide information about vegetation and ecosystems in the past. Since the isotope results from deer and wild boar from the Nukdo site fell within the range of C3 consumers, this means that the vegetation of the Nukdo Island and southern part of the Korean peninsula in the late Mumun period was composed of predominately C3 plants. These faunal data are consistent with other animal isotope data from neighbouring regions. Compared with data from the contemporary period in western Japan, there is not a large difference between the isotope data of the late Jomon and early Yayoi deer and wild boar (Minagawa et al., 2005; Mihara et al., 2004; Minagawa, 2001; Minagawa and Akazawa, 1992). For example, isotope average values of the late Jomon deer (n ¼ 9) from the Awazu site, Honshu, Japan, are d13C ¼ 23.0  1.2& and d15N ¼ 4.2  0.8&, and for the wild boar (n ¼ 9) from the Ikiriki site, Kyushu, Japan, are d13C ¼ 22.2  1.6& and d15N ¼ 4.3  0.8& (Minagawa et al., 2005). The Mumun period is regarded as a period of complex society with intensive agriculture. Isotope results of human remains from the Nukdo site indicate that there was a difference in the diets associated with gender. Females (10.9  0.4&) in the Nukdo site are approximately 1& lower d15N than males (11.7  1&), implying less consumption of protein resources. These isotope data reflect the variability in food consumption among individuals in the Nukdo site community. Overall average isotope values of human remains are 16 16 14 15 12 14 10 δ15NAIR δ15NAIR 18 8 Nukdo Human Wild Boar Deer Rat Dog Sea lion 6 4 2 Porgy Mullet Bear Bird Wild duck Cattle Infants & Children Female Mean Male Mean All adult Mean 13 12 11 10 9 0 -22 -21 -20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 δ13C -9 -8 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Age (years) vPDB 15 Fig. 2. Human and animal d13C and d15N values from the Nukdo site. (Nukdo human n ¼ 48; wild boar n ¼ 17; deer n ¼ 16; dog n ¼ 3; the other species n ¼ 9). Fig. 4. Juvenile collagen d N values plotted against age at death from the Nukdo site. The lines are the mean female and male adult value. Ages were determined by Kim et al. (1990, 1988). K. Choy, M.P. Richards / Journal of Archaeological Science 36 (2009) 1312–1318 -16 δ 13CvPDB -17 -18 -19 Female Mean Male Mean All Adult Mean Infants & Children -20 1317 humans from the late Mumun to early Iron Age periods of Korea. There was a mixture of both marine and terrestrial proteins in the diets, and marine protein was an important, although minor, food resource in these periods. The isotope data have also shown that there was no indication for the significant input of C4 plants into human diets. There is only isotope evidence of C3 plants, meaning that the Nukdo people primarily consumed C3 plants. It could be associated with the spread of rice agriculture in the Mumun period. The isotope data from the juveniles indicate that weaning occurred before the age of 1.5 years at the Nukdo site. Acknowledgements -21 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Age (years) 13 Fig. 5. Juvenile collagen d C values plotted against age at death from the Nukdo site. The line is the mean female adult value. Ages were determined by Kim et al. (1990, 1988). d13C 2–3& and d15N 7–8& higher than terrestrial herbivores. Especially, 7–8& d15N shifts in the trophic level can be interpreted as the influence of marine food on the whole diet, as in terrestrial diets the shift is normally lower at 5–6&. Carbon and nitrogen isotope signals show that human diets were mixed, with the main dietary protein coming from terrestrial sources and a minor consumption of marine resources. Because this site is located in the middle of the Nukdo Island and is covered in a shell midden, one might assume that these humans consumed a lot of marine foods. However, the archaeological excavations indicated that deer and wild boar bones were the main animal species at the site, whereas large sea mammal and fish bones were only present in small amounts. The carbon and nitrogen isotope data support the faunal evidence and show that humans at this site mainly consumed terrestrial animals fed on C3 plants with only a minor contribution from marine foods. It is generally accepted that the Mumun culture is closely related to western Japanese society in the Yayoi period. If we compare our results to those reported for one of the western Japanese sites, the latest Jomon and early Yayoi people in the Ohtomo site had much higher amounts of marine protein in their diets (Mihara et al., 2004). Carbon and nitrogen isotope results from the Ohtomo site are d13C ¼ 15.2& and d15N ¼ 11.8&, which is 3& higher in d13C than the humans from the Nukdo site. On the basis of this difference, we conclude that the Nukdo people consumed less marine protein than contemporary western Japanese groups. The Nukdo site yielded an extraordinarily large amount of infant skeletons. This provided the opportunity to explore breastfeeding and weaning patterns in the late Mumun period. Stable nitrogen isotopes can be used to determine breastfeeding and weaning behavior in past populations (Fuller et al., 2003, 2006; Herring et al., 1998; Katzenberg and Pfeiffer, 1995; Richards et al., 2002; Schurr, 1997). Nearly all children aged less than 5 years have elevated d15N results compared to the adult female mean and this d15N enrichment is caused by breastfeeding. After the average age of 1.5 years, stable nitrogen isotope ratios start to decrease, reflecting the dietary pattern of weaning as the nitrogen isotope values gradually drop to the same levels as the adult females. At Nukdo, eight infants seem to have been fully weaned by 1.5–2 years, whereas two children consumed breast milk as late as 3–3.5 years (Figs. 4, 5). The majority of Nukdo children seem to have been fully weaned by the age of 1.5 years. 7. Conclusions Stable isotope analysis of humans and animals from the Nukdo site has provided us with new information about the diet of All bone samples were provided through the kind courtesy of Dr. Ok-Ryun Jeon, the Prime Curator of the Busan National University Museum, South Korea. We would also like to thank A. Weiske and S. Boesel for valuable help in sample preparation and measurements. This research was funded by the Max Planck Society. References Ambrose, S., 1993. Isotopic analysis of palaeodiets: methodological and interpretive considerations. In: Sandford, M.K. (Ed.), Investigation of Ancient Human Tissue: Chemical Analyses in Anthropology. Gordon and Breach, Langhorne, Pennsylvania, U.S.A., pp. 59–130. Bale, M.T., 2001. Archaeology of early agriculture in Korea: an update on recent developments. Bulletin of the Indo-pacific Prehistory Association 21, 77–84. Bocherens, H., Drucker, D., 2003. Trophic level isotopic enrichment of carbon and nitrogen in bone collagen: case studies from recent and ancient terrestrial ecosystems. International Journal of Osteoarchaeology 13, 46–53. Brown, T.A., Nelson, D.E., Vogel, J.S., Southon, J.R., 1988. Improved collagen extraction by modified Longin method. Radiocarbon 30, 171–177. Busan University Museum, 1989. The Nukdo settlements. Busan University Excavation Reports 13 [in Korean]. Busan University Museum, 2004. The Nukdo settlements. Busan University Excavation Reports 29 [in Korean]. Cannon, A., Schwarcz, H.P., Knyf, M., 1999. Marine-based subsistence trend and the stable isotope analysis of dog bones from Namu, British Colombia. Journal of Archaeological Science 26, 399–408. Chisholm, B.S., Nelson, D.E., Schwarcz, H.P., 1982. Stable carbon ratios as a measure of marine versus terrestrial protein in ancient diets. Science 216, 1131–1132. Crawford, G.W., Lee, G., 2003. Agricultural origins in the Korean peninsula. Antiquity 77, 87–95. Crawford, G.W., Shen, C., 1998. The origins of rice agriculture: recent progress in East Asia. Antiquity 72, 858–866. DeNiro, M.J., 1985. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806–809. Fuller, B.T., Fuller, J.L., Harris, D.A., Hedges, R.E.M., 2006. Detection of breastfeeding and weaning in modern human infants with carbon and nitrogen stable isotope ratios. American Journal of Physical Anthropology 129, 279–293. Fuller, B.T., Richards, M.P., Mays, S.A., 2003. Stable carbon and nitrogen isotope variations in tooth dentine serial sections from Wharram Percy. Journal of Archaeological Science 30, 1673–1684. Hedges, R.E.M., Reynard, L.M., 2007. Nitrogen isotopes and the trophic level of humans in archaeology. Journal of Archaeological Science 34, 1240–1251. Herring, D.A., Saunders, S.R., Katzenberg, M.A., 1998. Investigating the weaning process in past populations. American Journal of Physical Anthropology 105, 425–439. Hu, Y., Ambrose, S.H., Wang, C., 2006. Stable isotopic analysis of human bones from Jiahu site, Henan, China: implications for the transition to agriculture. Journal of Archaeological Science 33, 1319–1330. Kaneko, H., 2004. Analysis of blocking sampling from the Nukdo shell midden, I C location. Busan University Excavation Reports 29, 445–462 [in Korean]. Kaneko, H., Seo, Y.N., 2004. Reports of faunal remains from the Nukdo I C location. Busan University Excavation Reports 29, 458–478 [in Korean]. Kaneko, H., An, J.H., Seo, Y.N., 1990. Reports of faunal remains from the Nukdo settlements. Gaya Tong Shin, 19–20 [in Korean]. Katzenberg, M.A., Pfeiffer, S., 1995. Nitrogen isotope evidence for weaning age in a nineteenth century Canadian skeletal sample. In: Grauer, A.L. (Ed.), Bodies of Evidence: Reconstruction History through Skeletal Analysis. Wiley-Liss, New York, pp. 221–235. Kim, J.J., Ogata, T., Mine, K., Takenaka, M., 1990. Human skeletal remains from the Nukdo site Loc I C. Gaya Tong Shin 19–20, 135–139 [in Korean]. Kim, J.J., Ogata, T., Mine, K., Takenaka, M., 1988. Reports of Human remains from the Nukdo site. Gaya Tong Shin 17, 53–58 [in Korean]. Lee, C.H., 2004. Reports of imported artifacts in the Nukdo shell middens, the Nukdo settlements. Busan University Excavation Reports 29, 541–559 [in Korean]. 1318 K. Choy, M.P. Richards / Journal of Archaeological Science 36 (2009) 1312–1318 Lee-Thorp, J.A., 2008. On isotopes and old bones. Archaeometry 50, 925–950. Mihara, S., Miyamoto, K., Nakamura, T., Koike, H., 2004. 14C age determination for human bones during the Yayoi period – the calibration ambiguity around 2400 BP and the marine reservoir effect. Nuclear Instruments and Methods in Physics Research B 223–224, 700–704. Minagawa, M., Matsui, A., Ishiguro, N., 2005. Patterns of prehistoric boar Sus scrofa domestication, and inter-islands pig trading across the East China Sea, as determined by carbon and nitrogen isotope analysis. Chemical Geology 218, 91– 102. Minagawa, M., 2001. Dietary pattern of prehistoric Japanese populations inferred from stable carbon and nitrogen isotopes in bone protein. Bulletin of the National Museum of Japanese History, 333–356 [in Japanese]. Minagawa, M., Akazawa, T., 1992. Dietary patterns of Japanese Jomon hunter-fishergatherers: stable nitrogen and carbon isotope analyses of human bones. In: Aikens, C.M., Rhee, S.N. (Eds.), Pacific Northeast Asia in Prehistory: Recent Research into the Emergence of Hunter-Fisher-Gatherers, Farmers and SocioPolitical Elites. University of Washington Press, Seattle, pp. 59–67. Pechenkina, E.A., Ambrose, S.H., Xiaolin, M., Benfer Jr., R.A., 2005. Reconstructing northern Chinese Neolithic subsistence practices by isotopic analysis. Journal of Archaeological Science 32, 1176–1189. Rhee, S.N., Choi, M.L., 1992. Emergence of complex society in prehistoric Korea. Journal of World Prehistory 6, 51–95. Richards, M.P., Fuller, B.T., Molleson, T.I., 2006. Stable isotope palaeodietary study of humans and fauna from the multi-period (Iron Age, Viking and Late Medieval) site of Newark Bay, Orkney. Journal of Archaeological Science 33, 122–131. Richards, M.P., Pearson, J.A., Molleson, I.I., Russell, N., Martin, L., 2003. Stable isotope evidence of diet at Neolithic Catalhoeyuek, Turkey. Journal of Archaeological Science 30, 67–76. Richards, M.P., Mays, S., Fuller, B.T., 2002. Stable carbon and nitrogen isotope values of bone and teeth reflect weaning age at the Mediaeval Wharram Percy Site, Yorkshire, UK. American Journal of Physical Anthropology 119, 205–210. Richards, M.P., Hedges, R.E.M., 1999. Stable isotope evidence for similarities in the types of marine foods used by late Mesolithic humans at site along the Atlantic coast of Europe. Journal of Archaeological Science 26, 717–722. Richards, M.P., Hedges, R.E.M., 1998. Stable isotope analysis reveals variations in human diet at the Poundbury camp cemetery site. Journal of Archaeological Science 25, 1247–1252. Schoeninger, M.J., Moore, K., 1992. Bone stable isotope studies in archaeology. Journal of World Prehistory 6, 247–296. Schoeninger, M., DeNiro, M., 1984. Nitrogen and carbon isotopic composition of bone collagen from marine and terrestrial animals. Geochimica et Cosmochimica Acta 48, 625–639. Schoeninger, M., DeNiro, M.J., Tauber, H., 1983. Stable nitrogen isotope ratios of bone collagen reflect marine and terrestrial components of prehistoric human diet. Science 220, 1381–1383. Schurr, M.R., 1997. Stable nitrogen isotopes as evidence for the age of weaning at the Angel site: a comparison of isotopic and demographic measures of weaning age. Journal of Archaeological Science 24, 919–927. Schwarcz, H., Schoeninger, M., 1991. Stable isotope analyses in human nutritional ecology. Yearbook of Physical Anthropology 34, 283–321. Sealy, J.C., 2001. Body tissue chemistry and palaeodiet. In: Brothwell, D.R., Pollard, A.M. (Eds.), Handbook of Archaeological Sciences. John Wiley & Sons, Chichester, UK, pp. 269–279. Suzuki, T., Fujita, H., Choi, J.G., 2008. Brief communication: new evidence of tuberculosis from prehistoric Korea-population movement and early evidence of tuberculosis in Far East Asia. American Journal of Physical Anthropology 136, 357–360.