Spatial and temporal variation of dietary habits during the prehistory of the Balearic Islands as reflected by 14C, δδ15N and δδ13C analyses on human and animal bones Mark Van Strydonck Mathieu Boudin Anton Ervynck Jaime Orvay Herlinde Borms Mayurqa (2005), 30: 523-541 SPATIALAND TEMPORAL VARIATION OF DIETARY HABITS DURING THE PREHISTORY OF THE BALEARIC ISLANDS AS REFLECTED BY 14C, δδ15N AND δδ13C ANALYSES ON HUMAN AND ANIMAL BONES Mark Van Strydonck* Mathieu Boudin** Anton Ervynck*** Jaime Orvay**** Herlinde Borms**** RESUMEN: Las islas Baleares forman un archipiélago en el Mediterráneo occidental. Las islas han estado colonizadas por humanos desde el Neolítico. El objetivo de este estudio era recabar datos sobre la importancia de la cría de animales, la agricultura y la pesca en la economía alimentaria de las sociedades baleares prehistóricas. Los datos 14C, δ15N y δ13C del colágeno y del carbonato de animales domésticos y de humanos abarcan diferentes períodos culturales y situaciones geográficas de los asentamientos, a saber, zonas montañosas, tierras bajas y litoral. En general, los datos isotópicos no corroboran las teorías tradicionales sobre la evolución de las estrategias de subsistencia durante la prehistoria balear. Dichos datos parecen indicar de nuevo que todos los períodos se caracterizan por una dieta mixta de productos animales y vegetales, y todo da a entender que las comunidades prehistóricas baleares no aprovechaban de un modo sistemático los alimentos marinos o de agua dulce. En más de un período se podrían haber dado pequeñas diferencias regionales en la dieta. PALABRAS CLAVE: Balear, colágeno, isótopo, animales, humanos, estrategias de subsistencia. ABSTRACT: The Balearic Islands are an archipelago in the western Mediterranean Sea. The islands have been occupied by humans since the Neolithic. The aim of this study was to garner information on the importance of animal husbandry, agriculture and fishing within the food economy of Balearic prehistoric societies. 14C, δ15N and δ13C data from collagen and carbonate of domestic animals and humans cover different cultural periods and a range of settlement locations, i.e., mountain, lowland and coastal sites. In general, traditional theories about the evolution of the subsistence strategies during the Balearic prehistory are not corroborated by the isotope data. It is again suggested that a mixed diet of animal and plant resources was characteristic for all periods, while it also * Royal Institute for Cultural Heritage, Jubelpark 1, B-1000 Brussels, Belgium. Corresponding author. Email: <[email protected]>. ** Royal Institute for Cultural Heritage, Jubelpark 1, B-1000 Brussels, Belgium. *** Flemish Heritage Institute, Phoenix-building, Koning Albert II-laan 19 box 5, B-1210 Brussels, Belgium. **** D.A.M.A.R.C., Deià, Mallorca, Spain. 525 Mark Van Strydonck et al. becomes apparent that prehistoric Balearic communities did not systematically use marine or freshwater food items. Small regional differences in diet may have occurred in more than one period. KEYWORDS: Balearic, collagen, isotope, animals, humans, subsistence strategies. INTRODUCTION The Balearic Islands form an archipelago in the western Mediterranean Sea, 80 to 300 km east of the Spanish south-eastern coast (figure 1.1). The islands have been colonised by humans, at least since the Neolithic (for detailed information on the different archaeological chronologies proposed for the Balearic Islands, see Waldren 1986, Plantalamor Massanet & Benejam 1997, Guerrero Ayuso 2001, Lull et al. 2002, Ramis et al. 2002). Since the 1960s, and still repeated in recent publications, the hypothesis has been launched that the subsistence strategy of the late Bronze Age - early Iron Age culture focused on animal husbandry, while plant cultivation remained of minor significance (see Hernandez-Gasch et al. 2002, and the references there). During the following period cereal agriculture would have gained importance (Mayoral Franco 1984). On the other hand, the Neolithic to early Bronze Age period would have been characterised by a mixed farming economy (Lewthwaite 1985). These reconstructions, however, are based upon a set of evidence that is not proof to criticism. Especially the almost ‘exclusive’animal husbandry system of the late Bronze Age - early Iron Age period has been questioned because it is clear that the cultural archaeological data could be biased due to preservation conditions, other taphonomic factors, excavation techniques, the choice of sites (being almost always ritual places), etc. (Hernandez-Gasch et al. 2002). Moreover, the environmental archaeological record for the Balearics is still very limited (see, e.g., the overview for animal remains in Chapman & Grant 1997). The analysis of plant and animal remains from archaeological sites can reveal which species were used for consumption and how the exploitation strategies were organised. Generally, however, it is very difficult, if not impossible, to evaluate the relative importance of animal versus plant products. Not only in sites with a hypothetical 100% survival of organic remains this would still be a methodological puzzle, in areas with less favourable preservation conditions, this is completely hopeless. The scarcity of plant remains from prehistoric Balearic sites can, for example, not be used in any interpretation about food patterns, since the phenomenon could be the result of taphonomic factors, of problems with sampling and recovery (given the state of excavation methodology in the area) or of the (ritual) nature of the sites. The preliminary archaeozoological data indicate that, at least at Son Ferrandell, the rearing of animals for meat alone was not the major aim of the site’s inhabitants, suggesting that vegetable food may have made the main contribution to the diet (Chapman & Grant 1997, 76). The same conclusion was reached by the analysis of trace elements from 24 bone samples from Càrritx (Pérez Pérez et al. 1999). To gain importance, however, these interpretations should be corroborated by an independent line of evidence. Moreover, there are other important questions, such as whether the absence of fish remains from the sites is the result of preservation conditions, of the lack of sieving during excavation, or of the subsistence strategy of the prehistoric people? An alternative approach for dietary reconstructions consists in the analysis of the isotopic fractionation of collagen and carbonate from the bones found in human burials from the period under study. It is well known that the δ13C and δ15N values in the body reflect the diet of an organism, a phenomenon that is now with increasing frequency used 526 Spatial and temporal variation of dietary habits during... for dietary reconstructions within archaeology (Sealy 2001). Very generally, the collagen of a bone is considered to reflect the isotope ratios of the protein fraction within the food, while the carbonate in the mineral fraction of the bone is considered to reflect the isotope ratios of the whole diet (van Klinken et al. 2000). This study presents data recorded from human bones from prehistoric sites from Mallorca, Menorca and Formentera. The aim is to see whether information about the subsistence strategy of these people could be extracted from the isotopic signatures, and whether diachronic trends were observable from the material. Isotope measurements on animal bones, often from the same sites from which human bone was studied, were taken as reference for the human data. The present report builds further upon a previous one (Van Strydonck et al. 2002a) but includes an important number of new data, and an new analysis method (isotope measurements from bone carbonate). Isotope data from archaeological material from the Balearic Islands have also been gathered by others (e.g., Davis 2002) but, for methodological reasons, only the dataset compiled by our own research will be used here. Alarge number of the radiocarbon dates presented here have already been listed by Van Strydonck et al. (1998, 2001, 2002b, 2004). MATERIALANDMETHODS The δ13C and δ15N data in this paper are a by-product of the radiocarbon analyses on animal and human bone from different sites at the Balearic Islands, performed, since 1986, at the Royal Institute of Cultural Heritage (Brussels, Belgium). This has both advantages and disadvantages: sample selection is relatively unbiased because all sites are potentially studied, but, on the other hand, not all samples dated in the past 18 years could be incorporated in the present study. For instance, in the beginning, stable isotope measurements were not taken, while the evaluation of the ∆δ13C is certainly an coll-carb analysis that was only performed during the later part of our investigation. 14C activity was measured using routine LSC (Forest and Van Strydonck 1995) or AMS (Van Strydonck and van der Borg 1991) procedures. The results are expressed as ages BP (Stuiver and Pollach, 1977). The pre-treatment of the bones followed Longin (1971). The protein content was checked by means of the C/N ratio by a Carlo Erba NA1500 analyser. All bone samples used in this study have a C/N 3, except IRPA-1179, ≤ IRPA-1066, UtC-9018, KIA-15220, KIA-20208, KIA-20461 with C/N values between 3.1 and 5.3. None of these samples, however, produced aberrant isotope data. Stable isotope measurements were performed on a Finnigin Mat Delta E mass spectrometer with a reproducibility of 0.1‰ for δ13C and 0.3‰ for δ15N. For certain animal bones, a species identification was not available, because the specimens had been subjected to radiocarbon dating before they were studied by an archaeozoologist. When identification was made, we were always dealing with domestic mammals, i.e. sheep, goat, cattle or pig. As traditional in archaeozoology, the remains of sheep and goat could not always be separated. Based on the local cultural entities (see the references above), the human samples can be subdivided in four chronological groups. These time periods coincide roughly with the Neolitic – Early Bronze Age period (phase 1: before 1600 BC), the Bronze Age (phase 2: 1600 – 1050 BC), the Iron Age (phase 3: 1050-500 BC), and a protohistoric to Early Roman period (phase 4: after 500 BC). 527 Mark Van Strydonck et al. RESULTS Table 1 and figure 1.2 give all data recorded from the sample set. The results will be discussed separately for the animal and the human sub-samples. Animal bones Figure 1.2 shows the clear difference between the scatter of animal samples and that of the human material. All together, the data for the animals show a large variation, a trend that must, at least partly, be explained by inter-species differences. Where species identifications of the animal samples were available, it became clear that most of the material sampled comes from true herbivores, i.e., cattle, sheep or goat, except for the single skeletal element of pig, and one of a dog (see further). Most of the unidentified animal samples will thus most probably also come from herbivores. Moreover, as far as we know, pigs were relatively rare in Balearic prehistory (Chapman & Grant 1997, 82, Figure 7.6). In general, the variation of the isotope fractionation within the group of herbivores will have been influenced by differences in the plants they consumed.Not only different plant species can have different isotope characteristics (see the references in van Klinken et al. 2000, 43), but also the spot where they grow can be highly influential. Altitude, salinity, local humidity and amount of sunlight (canopy effect) can have an influence on the δ13C of plants (see the references in Bocherens 2000, 73) and therefore also on the δ13C of the collagen of the herbivores. There are also differences in isotopic fractionation between different parts of the same plants (e.g., roots, leaves). The δ13C of the bone collagen of an animal will thus also be significantly determined by its foraging habits (Heaton 1999). Finally, it should be stressed that isotopic fractionation is largely different between two groups of plant species, having a different physiology, i.e., C3 and C4 plants. The latter group, however, was not present on the Balearic Islands in prehistory. In general, the different selection of plants, consumed by sheep, goat and cattle, will have caused the wide scatter of data points within the (mostly herbivorous) animal group. Furthermore, this pattern could explain the larger scatter of the goat-sheep datapoints compared to the cattle datapoints (two species versus one). That the sample of pig, an animal often described as omnivorous, falls within the variation of the true herbivores, is not really surprising since primitive herds of pigs, that are kept in natural environments, eat mostly plant material (in contrast to pigs that live in human habitations and are fed on consumption refuse) (Ervynck et al. 2003). Finally, a dog bone from Sa Cala (KIA-2015) showed stable isotope values that are comparable to those of the human bones, a pattern that can be explained by the omnivorous diet of a dog, comparable to that of many human populations, but different of that of real herbivores such as sheep, goat or cattle. Three distinct outliers are present within the animal dataset, one with an extremely negative 13C ratio, coming from the site of Ses Païsses (KIA-11890), whilst another, with an extremely high 13C ratio, comes from Son Gallard (KIA-23435). Athird animal outlier falls into the variation of the human samples: a goat jaw from Son Fornés (UtC-9327), of which the aberrant value is probably caused because a mixture of bone and tooth collagen was measured (see Bocherens 2000), The averages for the domestic mammals not taking into account the three aberrant samples and the data for the dog (δ13C = -20.41±0.75‰ and 528 Spatial and temporal variation of dietary habits during... δ15N = 5.44±1.77‰) coincide well with those listed in the literature for herbivorous mammals (δ13C= -21‰, δ15N= +5‰: Lanting and van der Plicht 1996). Finally, it has been investigated whether there is a diachronic trend in the measurements for the animal bone samples. However, no trend could be observed: regression analyses between the uncalibrated radiocarbon dates and δ13C or δ15N values proved that there were no relations between these variables (regression data not depicted here). Human bones When the total set of measurements for the human bones is compared with the animal samples, a general interpretation can be made about the former human diet. It is known that a predator has a less negative value of δ13C and a more positive value of δ15N compared to its prey. Theoretical studies suggest that the differences, describing the shift or a trophic level within the food chain, ideally should be +5‰ for δ13C (Ambrose 1993, but see further) and +2 to +4‰ for δ15N (Ambrose 1991), meaning that, when the human diet would have consisted solely of the meat of herbivores, these differences should describe the variation between the mean values for the isotope fractionations of humans and herbivores. Indeed, Lanting and van der Plicht (1996) give δ13C= -18‰ and δ15N= +8‰ for carnivores. When freshwater animals would have been added to the diet, more negative δ13C measurements would be obtained and more positive δ15N values. When marine food products would have been consumed, a less negative δ13C and a more positive δ15N would be recorded. When a human population would have survived solely on plant products, their isotope fractionation values must have been identical to those of herbivore mammals. In general, the isotope ratios of the human samples do not differ enough from that of the domestic animals to suggest a predominantly carnivorous diet (they certainly also do not attain the values listed by Lanting and van der Plicht 1996). Amixed diet of plant and animal material is thus much more likely. Generally, it must also be stressed that both N and C isotope fractionation values clearly indicate that the diet in Balearic prehistory was not predominantly based upon marine resources (implying that people living on an island were hardly fishing) and that people were also not concentrating upon the exploitation of freshwater animals (which can be explained by the islands’ biogeography). Would one of these strategies have been incorporated within the subsistence system, clearly different δ13C and δ15N values would have been obtained. Since the isotope data for the human material show a large variation, both for the δ13C and the δ15N ratios, it must now be established whether there are no significant differences between groups within the human sample population. Most likely, diachronic trends could be expected and therefore the data were split up taking into account the different phases defined earlier (figs 3 to 6). Figure 4.7 summarises the average isotope values for the animal and the four human datasets. The differences in δ13C values show no unidirectional diachronic trend and vary between 0.9 and 1.2‰. The differences between the average δ15N values of the animals and the human subgroups vary between 3.6 and 4.6‰. Disregarding the possible differences between individual sites (see further), the slightly higher δ15N value in the youngest phase could generally be explained by a change in diet compared to the earlier phases. This change, however, must have been rather limited because the differences of the δ13C and δ15N values observed between the four phases are certainly not dramatic. 529 Mark Van Strydonck et al. The inter-site differences seem much more important than the diachronic differences. This possibly suggests differences in diet between human populations on the islands or could be linked with the diverse ecology and geography of the islands, often showing marked differences between biotopes located within short distances. Moreover, most of the sites studied show specific structures, or characteristic burial rituals, thus suggesting different cultural (sub-)groups, a pattern that also makes acceptable that variation in subsistence strategy existed. These data also suggest that there is no important food trade between the different communities on the islands because this trade should mask the geographical differences. The data from the oldest human bones (phase 1: Figure 2.3) can be roughly divided in two groups: a larger group containing the samples from Alcaïdus, Biniai Nou, S’Aigua Dolça, Càrritx and Son Bauló and a second group containing the samples from Ca Na Costa and the cliff sites Cova Gregoria B, Son Moleta and Son Gallard. The two samples from Can Martorellet fall between the two groups. The isotope values of the second group can perhaps indicate a terrestrial subsistence strategy that was more carnivorous than that of the first group, or, in the case of sample KIA-14330 from Ca Na Costa, a diet that included a (limited) consumption of marine resources. The fact that there is no comparative material from Formentera, however, limits the interpretation possibilities. The samples representing phase 2 (Figure 2.4) show roughly the same picture. Once again the inter-site differences can be noted, although, in this phase, the differences between δ15N values are less outspoken. The human bone samples from phase 3 (Figure 3.5) show once again the same picture. Remarkable is the very tight group from Cova Gregoria A, except for one sample. This aberrant result can perhaps indicate that this particular individual was an immigrant from another community (Van Strydonck et al. in prep), or that weare dealing with a person that had a different diet because of his status, gender or even personal taste. Note that the same type of outliers, although less obvious, are also present in other sites and periods. The fact that the one sample from Cova Gregoria B is situated in the same part of the graph as those from cave A, indicates that the diachronic shift is much less important than the site depending variations. In phase 4 (Figure 3.6), however, one sample from Illa Des Porros shows a rather low δ15N value. Together with the sample from Biniai Nou, these specimens accentuate the larger variation in phase 4 compared to the preceding phases 2 and 3. Figure 4.8 summarises an additional approach within the analysis of the former dietary patterns on the Balearic Islands, i.e., the evaluation of the ‘spacing’between the δ13C values of the collagen and the carbonate fractions of the animal and human bones, in function of their date. It has been proven that this measurement yields considerably higher values for herbivores compared to carnivores. For the animal samples, this parameter varies around +10.5‰, a value considerably higher than listed for herbivores in the literature (+8 to +9‰: Bocherens 2000). How this difference must be explained, is not clear, but more important is that most of the values for the human samples do not really differ from those of the herbivorous domestic animals.In any case, the conclusion must be that a clear carnivorous diet cannot be established for a period within the pre-and protohistory of the Balearic Islands, but that differences between sites seem to exist. CONCLUSION The foregoing analysis represents a further attempt to use stable isotope analysis in order to make inferences about the diet of the pre- and protohistoric people inhabiting the 530 Spatial and temporal variation of dietary habits during... Balearic Islands. It has of course been impossible to exactly evaluate the relative importance of animal versus plant products within the diet, but the hypothesis of a mixed diet, relying on the exploitation of both resources, remains the most likely. Both isotope (δ13C and δ15N) values and spacing data between carbonate and collagen certainly yield no conclusive evidence for an exclusively carnivorous diet during a certain cultural period. Nevertheless, there are dietary differences between sites, but these are not systematically diachronic. During the oldest phase, for example, the difference between the data from the main islands and the small island of Formentera remains remarkable. During the youngest phase, slight differences in diet may have been present between sites, suggesting an increasing variation within diets. Finally, it is generally clear that marine or freshwater resources were hardly used by the prehistoric communities. REFERENCES AMBROSE, S. H. (1991): «Effects of diet, climate and physiology on nitrogen isotope abundances in terrestrial foodwebs». Journal of Archaeological Science18: 293-317. AMBROSE, S. H. (1993): «Isotopic analysis of palaeodiets: methodological and interpretive considerations». In: SANDFORD, M. K., editor: Investigations of ancient human tissue. Langhorne: Gordon & Breach Science Publishers. P. 59-130. BOCHERENS, H. (2000): «Preservation of isotopic signals (13C, 15N) in Pleistocene mammals». In: AMBROSE, S. H.; KATZENBERG, M. A.; editors: Biogeochemical approaches to palaeodietary analysis. New York: Kluwer Academic / Plenum Publishers. P. 65-88. CHAPMAN, R.; GRANT, A. (1997): «Prehistoric subsistence and monuments in Mallorca». In: BALMUTH, M. S.; GILMAN, A.; PRADOS-TORREIRA, L. editors: Encounters and transformations: the archaeology of Iberia in transition. Sheffield: Sheffield Academic Press. P. 69-87. DAVIS, MHLA(2002): «Putting meat on the bone: an investigation into palaeodiet in the Balearic Islands using carbon and nitrogen stable isotope analysis». In: Waldren, W. H.; Ensenyat, J. A. editors: World Islands in Prehistory. International Insular Investigations. V Deia International Conference of Prehistory. BAR International Series 1095. Oxford: Archaeopress. P. 198-216. ERVYNCK, A.; VAN STRYDONCK, M.; BOUDIN, M. (2003): Dieetreconstructie en herkomstbepaling op basis van de analyse van de stabiele isotopen 13C en 15N uit dierlijk en menselijk skeletmateriaal: een eerste verkennend onderzoek op middeleeuwse vondsten uit Vlaanderen. Archeologie in VlaanderenVII: 131-140. FOREST, L.; VAN STRYDONCK, M. (1995): The installation of a liquid scintillator and the construction of a benzene synthesis unit at the radiocarbon dating laboratory. Bulletin de l’Institut Royal du Patrimoine Artistique / Bulletin van het Koninklijk Instituut van het Kunstpatrimonium25: 235-245. GUERRERO AYUSO V. M. (2001): The Balearic Islands: Prehistoric colonization of the furthest Mediterranean islands from the mainland. Journal of Mediterranean Archaeology 14(2): 136-158. HEATON, THE (1999): Spatial, species, and temporal variations in the δ13C/δ12C ratios of C 3 plants: implications for palaeodiet studies. Journal of Archaeological Science26: 637-649. HERNANDEZ-GASCH, J.; NADAL, J.; MALGOSA, A.; ALESAN, A.; JUAN, J. (2002): Economic strategies and limited resources in the Balearic insular ecosystem: the myth of an indigenous animal farming society in the first millenium BC. In: Waldren, W. H.; Ensenyat, J. A., editors: World Islands in Prehistory. International Insular Investigations. V Deia International Conference of Prehistory. BAR International Series 1095. Oxford : Archaeopress. P. 275-291. 531 Mark Van Strydonck et al. LANTING, J. N.; VAN DER PLICHT, J. (1996): Wat hebben Floris V, skelet Swifterbant S2 en visotters gemeen? Palaeohistoria37-38: 491-519. LEWTHWAITE, J. G. (1985): Social factors and economic changes in Balearic prehistory, c. 3000- 1000 bc. In: Barker, G.; Gamble, C.; editors: Beyond domestication in prehistoric Europe. London: Academic Press. P. 205-231. LONGIN, R. (1971): New method of collagen extraction for radiocarbon dating. Nature230: 241- 242. LULL, V.; MICÓ, R.; RIHUETE, C.; RISCH, R. (2002). Social and ideological changes in the Balearic Islands during later prehistory. In: Waldren, W. H.; Ensenyat, J. A., editors: World Islands in Prehistory. International Insular Investigations. VDeia International Conference of Prehistory. BAR International Series 1095. Oxford: Archaeopress. P. 117-126. MAYORAL FRANCO, F. (1984): La fase post-talayotica Mallorquina: periodizacion y dinamica economico-social. In: Waldren, W. H. et al., editors. Early settlement in the Western Mediterranean islands and their peripheral areas. BAR International Series 229. Oxford: British Archaeological Reports. P. 1299-1313. PÉREZ PÉREZ, A.; FERNÁNDEZ, E.; TURBÓN, D. (1999): Paleodieta. In: Lull, V.; Micó, R.; Rihuette Herrada, C.; Risch, R., editors:Ideología y sociedad en la prehistoria de Menorca. La cova de Càrritx y la cova des Mussol. Ciutadella: Consell Insular de Menorca, Ajuntament de Ciutadella, Fundació Rubió Tudurí Andrómaco. P. 197. PLANTALAMOR MASSANET, L.; JUAN BENEJAM, G. (1997): Entra a la Menorca prehistòrica. Maó: Museu de Menorca. RAMIS, D.; ALCOVER, J. A.; COLL, J.; TRIAS, M. (2002): The chronology of the first settlement of the Balearic Islands. Journal of Mediterranean Archaeology15(1): 3-24. SEALY, J. (2001): Body tissue chemistry and palaeodiet. In: Brothwell, D. R.; Pollard, A. M., editors: Handbook of archaeological sciences. Chichester: John Wiley & Sons. P. 269-279. STUIVER, M.; POLACH, H. A. (1977): Discussion: reporting of 14C data. Radiocarbon19(3): 355- 363. VAN KLINKEN, G. J.; RICHARDS, M.; HEDGES, R. E. M. (2000): «An overview of causes for stable isotopic variations in past European human populations: environmental, ecophysiological, and cultural effects». In: Ambrose, S. H.; Katzenberg, M. A., editors. Biogeochemical approaches to palaeodietary analysis. New York: Kluwer Academic / Plenum Publishers. P. 39-63. VAN STRYDONCK, M.; BOUDIN, M.; ERVYNCK, A. (2002a): «Stable isotopes (13C and 15N) and diet: animal and human bone collagen from prehistoric sites on Mallorca, Menorca and Formentera (Balearic Islands, Spain)». In: Waldren, W. H., Ensenyat, J. A., editors: World Islands in Prehistory. International Insular Investigations. VDeia International Conference of Prehistory. BAR International Series 1095. Oxford: Archaeopress. P. 189-197. VAN STRYDONCK, M.; LANDRIE, M.; BOUDIN, M.; GROOTES, P. M.; NADEAU, M. J.; SPARKS, R.; KEPPENS, E. (2002b): Royal Institute for Cultural Heritage Radiocarbon dates XVIII. Brussels: Royal Institute for Cultural Heritage. VAN STRYDONCK, M.; LANDRIE, M.; HENDRIX, M.; VAN DER BORG, K.; DE JONG, A. F. M.; ALDERLIESTEN, K.; KEPPENS, E. (1998): Royal Institute for Cultural Heritage Radiocarbon dates XVI. Brussels: Royal Institute for Cultural Heritage. VAN STRYDONCK, M.; LANDRIE, M.; MAES, A.; VAN DER BORG, K.; DE JONG, A. F. M.; ALDERLIESTEN, K.; KEPPENS, E. (2001): Royal Institute for Cultural Heritage Radiocarbon dates XVII.Brussels: Royal Institute for Cultural Heritage. VAN STRYDONCK, M.; VAN DER BORG, K. (1991): «The construction of a preparation line for AMS-targets at the Royal Institute for Cultural heritage Brussels». Bulletin de l’Institut Royal du Patrimoine Artistique / Bulletin van het Koninklijk Instituut van het Kunstpatrimonium23: 228-234. VAN STRYDONCK, M.; LANDRIE, M.; BOUDIN, M.; GROOTES, P. M.; NADEAU, M. J.; KEPPENS, E. (2004): Royal Institute for Cultural Heritage Radiocarbon dates XIX. Brussels: Royal Institute for Cultural Heritage, in preparation. 532 Spatial and temporal variation of dietary habits during... VAN STRYDONCK, M.; BOUDIN, M.; ERVYNCK, A. (2004): «Possibilities and limitations of the use of stable isotopes (13C and 15N) from bone collagen as an aid in migration studies. Proceedings of the NATO workshop (ARW979859) Impact of the environment on human migration in Eurasia, (St.-Petersburg, 15-18 november 2003), E. M. Scott Andrey Yu. Alekseev and Ganna Zaitseva (eds.). NATO Science Series, IV. Earth and Environmental Sciences, Kluwer Academic Publishers. Vol. 42, p. 125-135. WALDREN, W. H. (1986): The Balearic pentapartite division of prehistory. BAR International Series 282. Oxford: British Archaeological Reports. 533