Ceramica - Sober Et Al.

8/14/2019 Ceramica - Sober Et Al.

1/8

253

UDK 90312\15(4\5)633\634>575.17< 90312\15(4\5)633\634>574.91Documenta Praehistorica XXXV (2008)

Early herding practices revealed throughorganic residue analysis of pottery from the early Neolithicrock shelter of Mala Triglavca, Slovenia

Lucija {oberl1, Andreja ?ibrat Ga[pari;2, Mihael Budja2 and Richard P. Evershed1*1 Organic Geochemistry Unit, School of Chemistry, University of Bristol, Bristol, UK2 Department of Archaeology, Faculty of Arts, University of Ljubljana, Ljubljana, SI

*author for correspondence> [email protected]

Introduction

Organic residues survive in two principal forms inassociation with archaeological pottery, namely as:(i) surface residues appearing as visible residues onthe exterior or interior of vessels, and (ii) as ab-sorbed residues preserved within the vessel wall; in-visible to the naked eye. The first class ofresidues, occurs on the exterior surfaces of and cor-respond to either sooting derived from heating ona fire or seemingly rather rare instances of applieddecorations (Urem-Kotsou et al. 2002; Connan etal. 2004). Together with exterior sooting, the inte-

rior surface residues are probably the group of resi-dues most familiar to pottery analysts. These areoften carbonised, and presumed to be residues of

cooking failures, while post-firing treatments arealso seen, particularly as copious linings associatedwith Roman amphorae (Beck et al. 1989). Absorbedresidues are by far the most common in pottery andprobably the most widely occurring residue type.Analyses performed to date suggest that absorbedorganic residues survive in >80% of domestic cook-ing pottery assemblages worldwide.

Lipids residues of cooking and the processing ofother organic commodities have been found to sur-

vive in archaeological pottery vessels as componentsof surface and absorbed residues for several millen-nia. The most successful analytical approaches in-

ABSTRACT A collection of pottery from the early Neolithic site of Mala Triglavca was analysed withthe aim of obtaining insights into vessel use and early animal domestication and husbandry prac-tices in the Adriatic region. Total lipid extracts were submitted to gas chromatography (GC), GC-massspectrometry (GC-MS) and GC-combustion-isotope ratio MS (GC-C-IRMS) in order to obtain molecu-lar and stable carbon isotope signatures as the basis for determining the nature and origins of theresidues. The extracts were dominated by degraded animal fats. The majority (70%) of the total lipidextracts displayed intact triacylglycerol distributions attributable to ruminant adipose and dairy fats,which were subsequently confirmed through C16:0 and C18:0fatty acid 13C values.

IZVLEEK Izbor keraminih vzorcev iz zgodnje neolitskega najdia Mala Triglavca je bil analiziran

z namenom, da bi pridobili dodatne informacije o uporabi keraminih posod ter znaaju zgodnje i-vinoreje na Jadranskem prostoru. Lipidne ekstrakte vzorcev smo analizirali s pomojo plinske kroma-tografije (GC), plinske kromatografije sklopljene z masno spektrometrijo (GC/MS) in plinske kroma-tografije sklopljene s seigno masno spektrometrijo razmerij izotopov (GC-C-IRMS). V lipidnih eks-traktih so prevladovale ivalske maobe. V veini (70%) lipidnih ekstraktov so bile prisotne nespre-menjene trigliceridne distribucije, ki jih lahko pripiemo tolnim ter mlenim maobam preveko-valcev. Lipidni izvor je bil nadalje potrjen z 13C vrednostmi prostih maobnih kislin C16:0 in C18:0.

KEY WORDS Neolithic, Adriatic, organic residues, fatty acyl lipids, 13C values


2/8

Lucija {oberl, Andreja ?ibrat Ga[pari;, Mihael Budja and Richard P. Evershed

254

volve solvent extraction, then using a combinationof instrumental analytical techniques, including: hightemperature-gas chromatography (HTGC), GC/massspectrometry (GC/MS;Evershed et al. 1990) and GC-

combustion-isotope ratio MS (GC-C-IRMS;Evershedet al. 1994), to identify and quantify the componentsof the lipid extracts of such residues. Characterisa-tion of lipid extracts to commodity type is only pos-sible through detailed knowledge of diagnostic com-pounds and their associated degradation productsformed during vessel use and/or burial. An increas-ing range of commodities is being detected in pot-tery vessels, including animals products (e.g. Ever-shed et al. 1992; Copley et al. 2003), leafy vegeta-bles (Evershed et al. 1991; Evershed et al. 1994),specific plant oils (Condamin et al. 1976; Copley et

al. 2005a) and beeswax (Heron et al. 1994; Ever-shed et al. 1997; Regert et al. 2001).

Animal fats are by far the most common residueidentified from archaeological pottery, characterisedby high abundances of free fatty acids, particularlypalmitic (C16:0) and stearic acid (C18:0). Triacylglyce-rols (TAGs) are the major constituents of modernanimal fats, however, they are degraded to diacylgly-cerols (DAGs), monoacylglycerols (MAGs) and freefatty acids during vessel use and burial, such that inarchaeological pottery the free fatty acids tend to

predominate. This has been observed in numerouspottery vessels (Evershed et al. 2001) and verifiedthrough laboratory degradation experiments (e.g.Charters et al. 1997; Dudd and Evershed 1998;Evershed 2008). Precise assigning of the origins ofanimal fats is only possible through the use of com-pound-specific stable carbon isotope analysis. GC-C-IRMS allows the carbon stable isotope (d13C) valuesof individual compounds (within a mixture) to bedetermined. It has been previously observed thatthe 13C values for the principal fatty acids (C16:0

and C18:0) are crucial in distinguishing between diffe-rent animal fats, e.g. ruminant and non-ruminantadipose fats and dairy fats (Evershed et al. 1997a,Dudd and Evershed 1998), as well as in the identi-fication of the mixing of commodities (Evershed etal. 1999; Copley et al. 2001). Recently it has beendemonstrated that dairy products were importantcommodities in Early Neolithic at various archaeolo-gical sites throughout Europe and Near East, as illu-strated through the persistence of dairy fats in ar-chaeological pottery vessels (Copley et al. 2003;2005b; Evershed et al. 2008).

The aim of this investigation was to apply organicresidue analysis to prehistoric pottery from the Neo-

lithic rock shelter Mala Triglavca in order to deter-mine the nature and origin of preserved lipids andthereby provide new insights into food preparationand consumption of the inhabitants. As a conse-

quence of the wider interest in the use of rock shel-ters in the early Neolithic, the organic analysis of thepottery from this site offers an important opportu-nity to explore aspects of animal husbandry, particu-larly dairying.

Sites and samples

The Neolithic rock shelter site of Mala Triglavca issituated in the Dinaric karst in south-western Slo-venia. There is evidence that the site has been con-tinuously occupied from the Mesolithic until the Mid-

dle ages. Rock shelter sites in the region have most-ly been interpreted in two different ways: (i) as sea-sonal camps for hunters/shepherds, or (ii) as placesfor long-term settlement. Archaeozoological remainsdiscovered on the site mainly belong to domestica-ted animals (cattle, sheep, goat, dog) as well as wildanimals (wild boar, red deer, roe deer). A total of 36potsherds were selected for analysis from earliestNeolithic phase.

Materials and methods

Lipid analyses were performed using established pro-tocols which are described in detail in earlier publi-cations (Evershed et al. 1990; Charters et al. 1993b).Briefly, analyses proceeded as follows:

Solvent extraction of lipid residuesLipid analyses of potsherds involved taking c. 2gsamples from area of the the sherd that had beensurface-cleaned using a modelling drill to removeany exogenous lipids (e.g. soil or finger lipids dueto handling). The sub-samples were then ground to

a fine powder, accurately weighed and a knownamount (20mg) of internal standard (n-tetratriacon-tane) added, to enable determination of the lipidconcentration. The lipids were extracted with a mix-ture of chloroform and methanol (2:1 v/v). Follo-wing separation from the ground potsherd the sol-vent was evaporated under a gentle stream of nitro-gen to obtain the total lipid extract (TLE). Portions(generally one third aliquots) of the extracts werethen trimethylsilylated and submitted directly to ana-lysis by HTGC. Where necessary combined GC/MSanalyses were also performed on trimethylsilylated

aliquots of the lipid extracts to enable the elucidationof structures of components not identifiable on thebasis of HTGC retention time alone.


3/8

Early herding practices revealed through organic residue analysis of pottery from the early Neolithic rock shelter of Mala Triglavca...

255

Preparation of trimethylsilyl derivativesPortions of the total lipid extracts were derivatisedusingN,O-bis(trimethylsilyl)trifluoroacetamide (40l;70C; 60 min; Sigma-Aldrich Company Ltd., Gilling-

ham, UK) and analysed by HTGC and GC-MS.Saponification of total lipid extractsMethanolic sodium hydroxide (5% v/v) was addedto the TLE and heated at 70C for 1 h. Followingneutralisation, lipids were extracted into chloroformand the solvent reduced under gentle stream of ni-trogen.

Preparation of methyl ester derivatives(FAMEs)FAMEs were prepared by reaction with BF3-metha-

nol (14% w/v; 100ml; Sigma-Aldrich, Gillingham,UK) at 70C for 1 h. The methyl ester derivativeswere extracted with chloroform and the solvent re-moved under nitrogen. FAMEs were re-dissolved inhexane for analysis by GC and GC-C-IRMS.

Results

The HTGC and GC/MS analyses serve to quantifyand identify compounds in the TLE, revealing thepossible presence of: (i) animal fat or plant oil, and/or (ii) plant epicuticular waxes, and/or (iii) beeswax

or other sealants, and/or (iv) mid-chain ketones thatindicate that the vessel has been heated (Evershedet al. 1995; Raven et al. 1997). Further analyses byGC-C-IRMS analyses can distinguish between rumi-nant and non-ruminant adipose fats and dairy fatsby investigating the d13C16:0 and d13C18:0values. Tab-le 1 lists the sample designations, the concentrationsof lipids detected and the assignments of the broadcommodity groups present in individual sherdsbased on the molecular and isotopic compositionsof the components of the TLEs. Ten of the sherds

(28%) yielded significant concentrations of lipid (i.e.>5mgg1) sufficient for further analysis by GC-MSand GC-C-IRMS.

Figure 1 shows a typical partial gas chromatogramfor the TLE of sample 08MT, revealing the presenceof free fatty acids, with high abundances of C16:0and C18:0 components, mono-, di- and triacylglycerols(MAGs, DAGs, TAGs). The chromatogram also showspresence of odd carbon number saturated fatty acids,iso- and anteiso-branched odd carbon number fattyacids (C15:0br, C17:0br), which may indicate a ruminant

source (Mottram et al. 1999; Evershed et al. 2001).Traces of wax esters were also present eluting in theregion of the TAGs.

MAGs, DAGs and TAGs, which are indicative of de-graded animal fat were detected in 7 of the TLEs to-gether with relatively high abundances of the pal-mitic (C16:0) and stearic (C18:0) free fatty acids, which

as discussed above, are the terminal products of TAGhydrolysis. Previous work has shown that the TAGdistributions can be linked to different sources,thereby allowing preliminary differentiation of fatsof the two major classes of domestic animals (rumi-nant and non-ruminant) and ruminant dairy fats.

However, laboratory experiments have shown thatTAG distributions can be skewed by degradation; thewide TAG distribution characteristic of fresh rumi-nant dairy fat is considerably narrowed, and thuscomes to resemble the narrower distribution seen in

ruminant adipose fat (Dudd et al. 1998; Dudd andEvershed 1998). Therefore conclusions drawn fromTAG distributions have to be made with caution. TheTAG distributions preserved in the extracts from theMala Triglavca sherds are shown in Figure 3.

The total lipid extracts (TLEs) of samples 08MT,18MT, 78MT, 79MT and 159MT displayed relativelybroad TAG distributions with acyl carbon numberrange of C44 to C54, maximising at C50/C52. Such dis-tributions are characteristic of reference ruminantadipose fat, or degraded milk fat. In contrast, the ex-

tract of 13MT displayed quite a narrow TAG distribu-

Fig. 1. Partial HTGC profile of the trimethylsilylatedtotal lipid extract from sample 08MT, showing thedistribution of components characteristic of degra-ded animal fat. Key: CX:0are saturated free fattyacids of carbon length x, br stand for branched

fatty acids, IS is the internal standard (C34 n-alka-ne). MAGs are monoacylglycerols; DAGs are dia-cylglycerols; TAGs are triacylglycerols.


4/8


256

tion, with an acyl carbon number range of C50 to C54,maximising at C52, which is identical to reference ru-

minant adipose fats (Fig. 2).

The 8 TLEs that yielded appreciable lipid concentra-tions analysed further by GC-C-IRMS to determinethe d13C values for the major fatty acids (C16:0 andC18:0); these values are plotted in Figure 4. The d13Cvalues obtained for modern reference animal fatsfrom the major domesticated animals exploited inprehistory are grouped within confidence ellipses,onto which the values from archaeological potteryare plotted. The d13C values for the C18:0 fatty acidare more depleted in milk fats than in ruminant adi-

pose fats, thus enabling distinctions to be drawn be-tween milk and adipose fats from ruminant animals(Dudd and Evershed 1998). This is witnessed in thec. 2.5 shift between centroids of the reference ru-minant adipose fat and ruminant dairy fat ellipses.The less depleted d13C values seen for the fatty acidsin non-ruminant fats compared to equivalent compo-

nents in ruminant fat are to be due to differences indiet and in the metabolic and biochemical processes

involved in the formation of body fats in ruminantand non-ruminant animals.

The d13C values for the C16:0 and C18:0 fatty acidsfrom 18MT, 79MT, 87MT, 88MT and 161MT plotwithin or adjacent to the dairy fat reference confi-dence ellipse, while that from 75MT plots within theruminant adipose reference fat ellipse. Values from08MT and 13MT plot between the porcine adiposefat and ruminant adipose fat ellipses. These d13C va-lues are most likely indicative of mixing of commodi-ties in the vessels, which may have occurred through

multiple use of the vessel or through the contempo-raneous mixing of animal products.

The modern fats used to construct the reference iso-tope plot were reared on a strict C3 diet of foddersand cereals. The slight displacement of some of thed13C values to the right of the mixing curves may be

Fig. 2. Histograms showing the typical acyl carbon number distributions expected for triacylglycerolsderiving from degraded lipid residues obtained from: (a) ruminant dairy fat, (b) ruminant adipose, and(c) pig adipose (Berstan 2002).

Fig. 3. The distributions of TAGs detected in the Mala Triglavca total lipid extracts.


5/8

Early herding practices revealed through organic residue analysis of pottery from the early Neolithic rock shelter of Mala Triglavca...

257

Lab. Lipidd13C16>0 d13C18>0 Predominant

sample concentration Lipids detected 0.3 () 0.3 () commodity type

no. (mg g1)

08MT 173.35FA (1618< 14, 15, 15br,17, 17br,

26.5 29.9 mixture of animal fats19, 20), MAG, DAG, TAG, WE

09MT 1.90 n\a n\a n\a n\a



13MT 11.56FA (1618), MAG, A,

26.7 28.8 ruminant adipose fatOH, DAG, WE, TAG





18MT 88.09FA (1618< 14, 15, 15br,17, 17br, 18>1,

27.7 33.3 dairy fat19, 20, 21, 22, 23, 24), MAG, DAG, TAG

75MT 90.54FA (1618< 14, 15, 15br,17, 17br,

29.0 31.5 ruminant adipose fat18>1, 19, 20), MAG, DAG, TAG




79MT 27.23FA (1618< 14,17, 17br, 18>1, 19,

27.8 32.9 dairy fat20), MAG, OH, A, DAG, TAG, P








87MT 21.93FA (1618< 14,17, 17br,

27.3 32.2 dairy fat18>1, 19, 20), OH, A, P

88MT 9.93 FA (1618< 14, 20), P 27.0 31.9 dairy fat


156MT 10.06 FA (1618< 18>1), P n\a n\a |

157MT 4.06 n\a n\a n\a n\a

158MT 0.98 n\a n\a n\a n\a159MT 12.65 FA (1618), MAG, DAG, TAG, P n\a n\a dairy fat |

160MT 5.53 n\a n\a n\a n\a

161MT 43.81 FA (1618), MAG, DAG 29.5 34.1 dairy fat

162MT 2.77 n\a n\a n\a n\a

163MT 4.02 n\a n\a n\a n\a

164MT 2.47 n\a n\a n\a n\a

165MT 1.53 n\a n\a n\a n\a

166MT 4.67 n\a n\a n\a n\a

Tab. 1. Summary of the results of the organic residue analyses of Mala Triglavca early Neolithic pot-sherds.

Key: FA refers to free fatty acids, MAG to monoacylglycerols; DAG to diacylglycerols; TAG to triacylglyc-erols; A are n-alkanes, K are mid chain ketones, WE are wax esters, P are plasticizers and nd = nonedetected. Annotation 18:1 refers to the level of unsaturation and 17br to branched free fatty acids.


6/8


258

due to the fact that the animals in prehistory werereared on diets, which varied in d13C values compa-red to todays values today environmental influen-ces. D13C values (d13C16:0 d13C18:0) are also a use-

ful indicator of lipid origin where such variationsexist. Figure 5 displays the D13C values plottedagainst d13C16:0 values for the Mala Triglavca pot-sherd fatty acids. The ranges on the left side of thegraph are from the modern reference fats.

Using Figure 5, it was possible to more securely attri-bute the Mala Triglavca residues to their potentiallipid sources. Sample 13MT, which was plotted onFigure 4 in between the ruminant dairy and adipose

reference ellipse, can now be more accurately attri-buted to the latter, together with 75MT. Unfortuna-tely, the same could not be achieved for TLE of sam-ple 08MT, which remains plotted on the boarder oftwo ranges and most likely the consequence of mix-

ing of different types of fat duringthe pottery use.

Discussion

The lipid components of the orga-

nic residues preserved in the earlyNeolithic vessels from Mala Triglav-ca displayed reasonable preserva-tion given their age, with appreci-able TLEs being detected in 28% ofthe sherds analysed. The high de-gree of preservation overall wasalso reflected in the survival of acyl-glycerol components (MAGs, DAGsand TAGs) in a significant propor-tion (70%) of TLEs. Although some-

what later age the lipid residues

from Mala Triglavca show similarrate of recovery and mean concen-trations to those observed in early

Fig. 4. Scatter plot showing the 13C values of C16:0and C18:0 fatty acids prepared from total lipidextracts of Mala Triglavca potsherds. The values ofmodern reference fats are represented by confi-dence ellipses (1 standard deviation). Lines con-necting the ellipses represent theoretical 13C val-ues obtained through the mixing of these fats.

Fig. 5. A plot showing the difference between 13Cvalues (13C18:0 13C16:0) and 13C values obtai-ned from the C16:0fatty acids extracted from theMala Triglavca potsherds. The ranges for the mo-dern reference fats are plotted to the left of the dia-gram.

Fig. 6. A plot showing the correlation between success rate (which isnumber of TLEs with appreciable lipid concentration divided by to-

tal number of samples analysed) and mean lipid concentration forMala Triglavca and other regions with evidences for early Neolithicmilk use.


7/8


8/8

COPLEY M. S., BERSTAN R., DUDD S. N., DOCHERTY G.,MUKHERJEE A. J., STRAKER V., PAYNE S. and EVERSHEDR. P. 2003. Direct chemical evidence for widespread dair-ying in prehistoric Britain.Proceedings of the NationalAcademy of Sciences 100: 15241529.

COPLEY M. S., BLAND H. A., ROSE P., HORTON M. andEVERSHED R. P. 2005a. Gas chromatographic, mass spec-trometric and stable carbon isotopic investigations of or-ganic residues of plant oils and animal fats employed asilluminants in archaeological lamps from Egypt.Analyst130: 860871.

COPLEY M. S., BERSTAN R., DUDD S. N., AILLAUD S., MUK-HERJEE A. J., STRAKER V., PAYNE S. and EVERSHED R. P.2005b. Processing of milk products in pottery vesselsthrough British prehistory.Antiquity 79: 895908.

DUDD S. N. and EVERSHED R. P. 1998. Direct demonstra-tion of milk as an element of archaeological economies.Science 282: 14781481.

EVERSHED R. P. 1993. Biomolecular archaeology and li-pids. World Archaeology 25: 7493.

2008. Experimental approaches to the interpretationof absorbed organic residues in archaeological ceram-ics. World Archaeology 40: 2647.

EVERSHED R. P., HERON C. and GOAD L. J. 1990. Analysisof organic residues of archaeological origin by high-tem-perature gas-chromatography and gas-chromatographymass- spectrometry.Analyst 115: 13391342.

1991. Epicuticular wax components preserved in pot-sherds as chemical indicators of leafy vegetables in an-cient diets.Antiquity 65: 540544.

EVERSHED R. P., HERON C., CHARTERS S. and GOAD L. J.1992. In A. M. Pollard (ed.),New Developments in Archa-eological Science. Oxford University Press, Oxford: 187208.

EVERSHED R. P., ARNOT K. I., COLLISTER J., EGLINTON G.and CHARTERS S. 1994. Application of isotope ratio mo-nitoring gas-chromatography mass-spectrometry to theanalysis of organic residues of archaeological origin.Ana-lyst 119: 909914.

EVERSHED R. P., VAUGHAN S. J., DUDD S. N. and SOLESJ. S. 1997. Fuel for thought? Beeswax in lamps and conicalcups from the late Minoan Crete. Antiquity 71: 979985.

EVERSHED R. P., DUDD S. N., CHARTERS S., MOTTRAM H.,STOTT A. W., RAVEN A., VAN BERGEN P. F. and BLAND H.A. 1999. Lipids as carriers of anthropogenic signals from

prehistory.Philosophical Transactions of the Royal So-ciety of London Series B-Biological Sciences 354: 1931.

EVERSHED R. P., DUDD S. N., COPLEY M. S., BERSTAN R.,STOTT A. W., MOTTRAM H. R., BUCKLEY S. A. and CROS-SMAN Z. 2001. Chemistry of archaeological animal fats.Accounts of chemical research 35: 660668.

EVERSHED R. P., PAYNE S., SHERRATT A. G., COPLEY M. S.,COOLIDGE J., UREM-KOTSOU D., KOTSAKIS K., ZDOGANM., ZDOGAN A. E., NIEUWENHUYSE O., AKKERMANS P.M. M. G., BAILEY D., ANDEESCU R. R., CAMPBELL S., FA-RID S., HODDER I., YALMAN N., ZBASARAN M., BIAKCIE., GARFINKEL Y., LEVY T. and BURTON M. M. 2008. Ear-liest date for milk use in The Near East and southeasternEurope linked to cattle herding.Nature 455: 528531.

HERON C., NEMCEK N. and BONFIELD K. M. 1994. Thechemistry of Neolithic beeswax.Naturwissenschaften 81:266269.

MLEKU D. 2006.Meat or milk? Neolithic economies ofCaput Adriae. Preistoria dellItalia settentrionale. Studiin ricordo di Bernardino Bagolini, Atti del Convegno, Udi-ne settembre 2005, 453 458.

MLEKU D., BUDJA M., PAYTON R. and BONSALL C. 2008.Mind the Gap: Caves, Radiocarbon Sequences, and theMesolithic-Neolithic Transition in Europe Lessons from

the Mala Triglavca Rockshelter Site. Geoarchaeology: AnInternational Journal 23: 398416.

MOTTRAM H. R., DUDD S. N., LAWRENCE G. J., STOTT A.W. and EVERSHED R. P. 1999. New chromatographic,mass spectrometric and stable isotope approaches to theclassification of degraded animal fats preserved in archa-eological pottery. Journal of Chromatography A. 833:209221.

RAVEN A. M., VAN BERGEN P. F., STOTT A. W., DUDD S.N. and EVERSHED R. P. 1997. Formation of long chain ke-tones in archaeological pottery by pyrolysis of acyl lipids.Journal of Analytical and Applied Pyrolysis 40/41: 267285.

REGERT M., COLINART S., DEGRAND L. and DECAVALLASO. 2001. Chemical alteration and use of beeswax throughtime: accelerated ageing tests and analysis of archaeologi-cal samples from various environmental contexts.Archa-eometry 43: 549569.

UREM-KOTSOU D., STERN B., HERON C. and KOTSAKIS K.2002. Birch-bark tar at Neolithic Makriyalos, Greece.An-tiquity 76: 962967.


260

Ceramica - Sober Et Al.

Documents

Transcript of Ceramica - Sober Et Al.