Cramp, L. J. ., Evershed, R. P., & Eckardt, H. (2011). What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain. Antiquity, 85(330), 1339–1352 . http://antiquity.ac.uk/ant/085/ant0851339.htm

Publisher's PDF, also known as Version of record

Link to publication record in Explore Bristol Research PDF-document

University of Bristol - Explore Bristol Research General rights

This document is made available in accordance with publisher policies. Please cite only published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/red/research-policy/pure/user-guides/ebr-terms/ h oa ceramic Roman The Introduction Cramp J.E. Britain Lucy Roman and Age Iron in change cultural and residues for? Organic used mortarium a was What ihrGli egc rsuhesenBian(ikno ate 91.B h later the By 1971). in other Hartley the & and (Dickinson Albans) Britain (St Verulamium south-eastern of or south Belgica situated Gallia was dominate to which either began of centres one major two market, locally-made and the and Britain imported in both common 43, increasingly low, AD became initially After mortaria was trade. trade larger-scale of other scale to the that incidental (Hartley suggests perhaps artefacts south-east these the of in frequency predominantly The Age, 1985). located Iron 1981, sites later the at to discovered date imports which continental Britain, from are recovered mortaria earliest The spout. and rim eevd 8Nvme 00 cetd 0Jnay21;Rvsd ac 2011 March 7 Revised: 2011; January 30 Accepted: 2010; November 18 Received: Keywords: 2 1 ANTIQUITY Ealfrcrepnec:[email protected]) correspondence: for (Email rai eceityUi,Sho fCeity nvriyo rso,CnoksCoe rso S T,UK 1TS, BS8 Bristol Close, Cantock’s Bristol, of University Chemistry, UK of 6AB, School RG6 Unit, Reading Geochemistry Whiteknights, Organic Reading, of University Archaeology, of Department 5(01:13–32http://antiquity.ac.uk/ant/085/ant0851339.htm 1339–1352 (2011): 85 rnAeadRmnBian rtfut etre D otra it lipids diet, mortaria, AD, centuries first–fourth Britain, Roman and Age Iron mortarium 1 , 2 ihr .Evershed P. Richard , sarbs, omo olo ai,wt prominent a with basin, or bowl of form open robust, a is 1339 rfc creams? face or pharmaceuticals food, for they Were mortaria. in together mixed these being were why products natural is The however, remains, products. that dairy question the contained including mortars in fats, the cooked plants, animal as as well well As as pot. ground preparing mortarium being the of were in method plants products: the that certain just diet the — wasn’t it changed that show pots, authors cooking the Roman and Age Iron with them the comparing in and blenders’ ‘kitchen trapped these of residues walls purpose. the actual of its analysis Using on to speculation has there been mortarium and long AD, century the first the in brought Britain Romans The 2 Eckardt Hella & 1

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain

Figure 1. Drawing of a typical mortarium found in Roman Britain.

first century, however, imports had all but ceased, with only very low numbers of Gaulish mortaria present after AD 100 (Hartley 1973). Production was no longer confined to south- east Britain, with many other industries in the Midlands and further north producing and distributing mortaria alongside the still-dominant south-eastern production centres. From the later second century mortaria became more frequent and their distribution spread from a predominantly urban and military distribution to rural settlements. At rural sites in the north and west of the province of Britannia they are often found to comprise a relatively significant proportion of the overall ceramic assemblage (Evans 1995; Tyers 1996; Rush 1997; Cool 2004, 2006) even if the overall contribution of Roman-type pottery remained low. The earliest mortaria in Britain were straight (wall)-sided, often with incised lines scored around the interior providing a roughened surface (Hartley 1981, 1985; Parminter & Hartley 1996). These examples were usually not spouted. However, by the mid first century this version became obsolete, being replaced by the classic mortarium exhibiting a well- defined rim and flange, with coarse trituration grits replacing the internal scoring (Figure 1). Regional or industry-specific styles developed over time, including the Mancetter-Hartshill ‘hammer-head’ form and the Oxfordshire colour-coated wares imitating earlier Gaulish Samian ware imports. Although distinctive in form, the size, fabric and style of mortaria could vary widely; whilst most fall within the range of 20–30cm in diameter, miniature vessels and examples reaching 1m in diameter are also known (Phelps 1923). Spouts, when present, could flat and broad, a mere finger impression on the rim or a pierced hole through the vessel wall. Decoration was uncommon, but painted, stamped, incised or rouletted examples are known (Young 1973, 1977). Heavy wear has been observed on some sherds (Hartley 1981, 1990; Parminter & Hartley 1996), which, combined with the presence of the trituration grits, indicate that the vessels may have been subjected to considerable abrasive activity. Sooting or burning is

1340 w rts ie n rnAecoigwrsfo orBiihsts(al 1). (Table sites in British four wares from cooking wares (six Roman cooking sites from Age seven Iron residues from and with mortaria sites compared Roman British were in two residues continental) The one vessels. and the British residues of of fabric characterisation of existing the the types in through of the mortaria, preserved establish service in 1923; to processed (Phelps the undertaken being study processing were to a cereal that reports put commodities or paper utensil 1943) This (Oswald 2001)? new Alcock cheese a 1973; making Hartley it as such was cuisine, or Age 1995), Iron Evans (e.g. preparation food etesae bet n ag fohrclnr bet n odies a eidentified be can items, (Bl food Malaga and from objects mosaic culinary alongside a other mortaria, in of of a range depiction grinding a in two-dimensional the and substances indicating A objects as pestle-shaped grinding 2). interpreted Figure or Museum; gesture origin, mixing (British latter Egyptian slave this onions — of of a figure, eye figurine, depicts seated an and similar wiping a Museum whilst A by British mortarium ambiguous. use the is at in contents exhibited mortarium the display is a of on depicts nature provenance This the unknown Oxford. although of Museum, Ashmolean figurine the Roman in a including artefacts, contemporary (Apicius, add liquor” vinegar, cooking with the 1958). moisten of mortar Rosenbaum some a pound; over in rue; pour “Put mint, dates, ingredients: Jericho root, culinary of of asafoetida pounding context coriander, or the caraway, mixing in pepper, to mortarium reference a with to refer often Columella preparations, and Apicius 1998). Dalby 75–6; fteidvda at cd rsn ntelpdrsde (Heron residues composition lipid isotopic to the carbon as in stable specification present the Further acids of vessels. fatty analysis the the individual through the of obtained of contents is original fats the of to origin ‘fingerprint’ the back a related providing be (GC/MS), may of spectrometry using that characterised fabric GC/mass and and the separated of (GC) from lipids walls of chromatography extracted mixtures gas the be these and in may solvents lipids absorbed organic Preserved using residues pottery established. organic well of are identification pottery and unglazed detection the for Methods Method such dishes prepare to order in wine, and sauce and fish sauces oil, rissoles, meat, as spices, and herbs 45–6; including 1943: (Oswald sites 565). of 1996: range May a 109; from 1999: sherds 194–5, on 1990: observed Hartley been has but ubiquitous not oprdwt rvosypbihddt rm27Io g esl (Copley vessels Age Iron 237 are from These data lipids. 15 of published from concentrations previously deriving appreciable vessels with containing Roman compared their 600 of basis some the from on selected assemblages, were here examined residues The Results 90 92 99 Dudd 1999; 1992, 1990, osteitouto ftemraimsga h dpino e Rmn tl of style ‘Roman’ new of adoption the signal mortarium the of introduction the Does vdnefrteueo otracnb eie rmvsa ersnain in representations visual from derived be can mortaria of use the for evidence Other codn oRmnsucs otrawr sdt oehrarneo ingredients, of range a together mix to used were mortaria sources, Roman to According uyJE rm,RcadP vrhd&HlaEckardt Hella & Evershed P. Richard Cramp, J.E. Lucy moretum tal et zuz1981). azquez ´ 99 Evershed 1999; . akn fces-ra;eg aoteElder, the Cato e.g. cheese-bread; of kind (a 1341 ? ). Coquinaria re tal et 91 Evershed 1991; . eAgricultura De tal et ;Flower& 2005). . tal et .

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain

Figure 2. Figure of a slave handling substances in a mortarium whilst rubbing his eye. Origins: Egypt, c. AD 200 (c The Trustees of the British Museum).

Despite the high number of vessels investigated, lipid recovery from mortaria was poor, such that only vessels from seven of these assemblages are considered further. Moreover, although the proportion of sherds containing residues from all types of vessel appears consistent, it was noted that absolute lipid concentrations recovered from mortaria were distinctly lower than both Iron Age and Roman ‘cooking’ vessels recovered from the same site (Figure 3). These low concentrations may be due to the predominant use of the mortarium to grind plants without heat: it has been shown that residues are more easily absorbed when heated, and animal tissues release higher concentrations of lipids than most plant tissues (Charters et al. 1993; Charters 1997; Evershed 2008). The plant-derived components that were identified in the mortarium sherds include ranges of long-chain odd-carbon number n-alkanes, even-carbon number n-alkanols,

1342 ise il ihrcnetaino iista ln ise Dd 99,sc that such 1999), (Dudd tissues plant over-represented. than quantitatively lipids usually is of processing concentration product animal higher exceptional a with were to yield observed 60 origin were tissues in plant residues occurring and mortaria, plant-derived animal Roman groups, in of vessel frequency fats the of of compared all mixtures When mortaria in Although in 4). identified waxes 5). plant (Figure (Figure of distinctive esters frequency clearly high wax the is vessels, intact ‘cooking’ domestic even Roman and other with sterols plant ketones, mid-chain atnWspai,Germany Westphalia, Xanten rxtrSrwbr Roman Shrewsbury Wroxeter tnaGag abigsieRoman Cambridgeshire Grange Stonea Stanwick irerdeC.Dra Roman Durham Co. Piercebridge ihoreW usxHg status High Sussex W. Fishbourne adnDre rnAehlfr 44 (76) 41 54 hillfort Age Iron Dorset Maiden Danebury (Tunisia). Carthage and Italy) from (Campania, vessels Sedgefield Pompeii 180 (Leics.), (Hamps.), included Villas Silchester these Rutland lipid Durham), but (Leics.), which (Co. table Leicester from this (Essex), Sites in Heybridge, discussed. included (Manchester), are not Deansgate residues are pottery poor which very from was sites preservation of Summary 1. Table n/a † ieRgo yeMrai esl characterised vessels Mortaria Type Region Site ano xodhr rnAe4 8(57) 28 49 rural Roman Age Iron Durham Co. Oxfordshire Faverdale Yarnton rnAedt ulse nCopley in published data Age Iron Castle Cresswell = analysed not † † † † otaposieIo g and Age Iron Northamptonshire aphr rnAehlfr 93 (48) 33 69 hillfort Age Iron Hampshire uyJE rm,RcadP vrhd&HlaEckardt Hella & Evershed P. Richard Cramp, J.E. Lucy tal. et 2005 Colonia Civitas and fortress legionary centre ministrative ad- ?imperial village agricultural Roman fort later and town villa/small ‘palace’ Roman farmstead settlement capital 1343 > 0 ftedansi eius Animal residues. diagnostic the of 90% ubro vessels of Number 1na3 (46) 33 n/a 71 0na1 (65) 13 n/a 20 8Io g:65 Age: Iron 38 0na2 (50) 20 n/a 40 9na2 (69) 20 n/a 29 6na4(15) 4 n/a 26 25 otra 0(46) 10 Mortaria: 55 22 analysed ‘Cooking’ oa:93 Roman: rnAecooking Age Iron n okn:3 (52) 31 cooking: Roman (66) 25 mortaria: Roman (60) 39 vessels: okn:4 (66) 41 cooking: ( % eslresidues vessel )

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain n-nonacosane was usually the major component of plant-derived lipid residues; however, this is a relatively ubiquitous component of plant lipids and further, one that is more likely to resist decay compared with alkanols and ketones (Char- ters 1997) or shorter-chain homologues (Regert et al. 2001). Other components present include n-alkanes within the range C27–C33, n-alkanols between C24–C32 and the mid-chain ketone nonacosan-15-one. Intact wax esters were detected in residues from Faverdale, Stanwick and Wroxeter; these were predominantly in the range C44– C52 with fatty acid moieties of carbon chain length C18–C24. The range and relative abundance of these plant-derived components was variable between and even within sites. Therefore, there is no reason to suspect that the plant material that was processed in mortaria was necessarily of a single type. Further, there is no discernible qualitative difference between the plant-derived components observed in the mortaria and those observed in Iron Age and Roman ‘cooking’ vessels. The concentration of C18:0 in plant tissues is very low and as such, the stable carbon isotopic composition could not be measured in the residues that were almost Figure 3. Box plots derived from the lipid concentrations entirely plant-derived. However, the stable from absorbed residues preserved in Roman pottery from carbon isotopic composition of individual Stanwick and Faverdale. fatty acids from the leaves, stems and seed oils of plants is undiagnostic of plant type, other than to separate C3 and C4 species (Woodbury et al. 1998; Dungait et al. 2008, 2010) and therefore such values would not enhance our determination of the source of plant-derived lipids. Biomarkers from plant resins were also detected in two mortarium sherds. The resins identified in sherds from Faverdale and Piercebridge are diterpenoid resins, from which the Pinaceae family (pine, cedar, fir) may be distinguished by the presence of components deriving from abietane and pimarane skeletons (Zavarin & Snajberk 1980; Evershed et al. 1985; Simoneit 1986; Serpico & White 2000). In these residues, the identification of abietic acid, pimaric acid and related degradation products, including dehydroabietic acid, indicates that Pinaceae resin was associated with the mortarium, and further, the presence of the methyl ester of dehydroabietic acid is indicative that the resin was heated (e.g. pitch). It is unlikely that the mortarium was used directly to prepare resin or pitch as the concentrations

1344 X DAGs iue4 ihtmeauegscrmtgaso yia oa ii xrcs(rmtyslltd bandfo hrsfrom sherds from obtained (trimethylsilylated) extracts Abbreviations: lipid text. total the typical in discussed of sites chromatograms seven gas the temperature High 4. Figure OH = = icllcrl;TAGs diacylglycerols; n aknlwith -alkanol X uyJE rm,RcadP vrhd&HlaEckardt Hella & Evershed P. Richard Cramp, J.E. Lucy abnatoms; carbon = raygyeos IS triacylglycerols; X A X: = = n FA akn with -alkane nenlsadr ( standard internal = 1345 reftyai with acid fatty free X abnatoms; carbon n -tetratriacontane); X abnaosadydge funsaturation; of degree y and atoms carbon X K = eoewith ketone ∗= hhlt contamination. phthalate X abnatoms; carbon

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain

Figure 5. Box plot showing the proportion of residues containing plant-derived components identified in the Iron Age and Roman assemblages. The analysis of Iron Age vessels produced 141 diagnostic residues from four sites: Danebury (Hamps.) Yarnton Cresswell Field (Oxon.), Maiden Castle (Dorset) and Stanwick, (Northants; Copley et al. 2005). The Roman ‘cooking’ vessel group includes 72 diagnostic residues from Stanwick and Faverdale (Co. Durham), whilst the mortaria sample group includes 125 diagnostic residues from six British and one continental site: Faverdale, Fishbourne (W. Sussex), Piercebridge (Co. Durham), Stanwick, Stonea (Cambs.), Wroxeter (Shrops.) and Xanten (Westphalia, Germany).

of diterpenoids are very low; however, it is a possibility that these biomarkers result from their dissolution in a non-viscous commodity that had been stored previously in amphorae with a resin lining. Pine resins have previously been chemically identified on amphorae believed to contain wine (Peacock & Williams 1986; Heron & Pollard 1988; Colombini et al. 2005) and recently as a pitch coating in a Dressel 20 amphora from London (Stern et al. 2008), a form widely believed to have been used to transport olive oil (Tyers 1996). Resin acids from such a lining would readily dissolve in olive oil and moreover would easily detach and become mixed with the contents; as such, biomarkers for the pitch lining may have been transferred into subsequent preparation vessels. Whilst plant components are widely detected in mortarium residues, degraded animal fats, characterised by a high relative abundance of C18:0 fatty acid (stearic acid) and distinctive distributions of intact triacylglycerols, are also frequently observed in residues from these vessels. The occurrence of animal-derived fats suggests that both animal and plant products were regularly prepared in the same mortaria, although it cannot be discerned whether these were mixed on the same occasions, or whether the mortarium was used separately for a variety of commodities. Pottery from Piercebridge contained unusually high concentrations of steroidal components, dominated by C27 sterols, stanols (5α and 5β epimers) and stanones, which are diagnostic of an animal origin. Plant-derived C28–C29 sterols and stanols, including campestanol, β-sitosterol and stigmastanol, were also identified in lower concentrations in the same residues. These findings agree with the other biomarker evidence,

1346 nmlft n ln is ntersligrsde.I sitrsigt oeta hr slittle is there that note e.g. to interesting content, is lipid It residues. absolute resulting higher the in a oils, with plant and commodities fats of animal representation the stage. towards this bias at special discounted a be residues. cannot of the specimens possibility in larger the the therefore fats for and animal function residues analysed cereal-related of less-diagnostic forms mortarium and frequency larger poor the the yielded that by observation generally the supported by qualified not be must is this However, grain of processing the a commonly of possibility more the were form. Even precluding vessel thereby waxes identified. this fats, for epicuticular dairy rarely use than plant unique otherwise residues and mortarium were the fats in origin adipose identified since routinely dairy vessels) ruminant cheese-making were predominantly derive were Stanwick, (e.g. a mortaria at products products to dairy from that dairy processing appeared deriving hypothesis for (Stanwick), fats vessel products the site specialised a reject animal one as used at The However, process although origin. processed. or carcass extensively contain animal the mortaria to and intensively from Roman or predominantly from plant frequently residues of used organic were commodities vessels absorbed these diagnostic that 125 demonstrates of interpretation The Discussion Dudd 1998; Evershed & (Dudd Evershed source 1999; dairy ruminant ruminant porcine), and (e.g. (sheep/goat/cow) non-ruminant predominantly carcass a product of fats of plant separation the and example, for animal both of importance mortaria. in the processing concerning conclusions supporting oee,tebodrrl fmrai per ossetwt ie lehr nBritain. in elsewhere products, sites with adipose consistent replace, appears mortaria or of supplement, role broader to the However, here availability ready used their to likely widely due seems components it perhaps were Therefore, lipid 3). products plant-derived & distinguished of dairy 2 easily (Figures frequency overall that be high lipid of still the concentrations to may lower and due residues products vessels mortarium pot the cooking frequency Stanwick, the high at from unusually residues the ( fat Despite fats dairy products. of butter dairy upon or dependence milk considerable of with proportion high a exhibited rmSawc eie rmapeoiatydiysuc.Itrsigy h eiusfrom residues (Copley the Interestingly, Age source. Iron dairy the predominantly both a 6). from (Figure derived pots Stanwick the from therefore in processing fats, fat adipose or cattle) meat or goat or goat sheep sheep, cattle, fats (e.g. of indicating majority (e.g. ruminant The non-ruminant contents. from that mortarium these predominantly of evidence origin component of originate little the significant is compositions a clarify there comprised isotopic further ever that fat to demonstrate carbon pig) findings order stable These in the fats. determined the and were of origin residues animal mortarium in of components fats in acids fatty sarltvl mliproe iigbw,i ol eepce htteewudb a be would there that expected be would it bowl, mixing ‘multi-purpose’ relatively a As for solely use a disproven, be cannot cereals grinding for mortaria of use the Whilst allowing fats, of source the reflect acids fatty individual of values isotopic carbon stable The lhuhdiyft eerr,treo e otru eiussbetdt GC/C/IRMS to subjected residues mortarium ten of three rare, were fats dairy Although tal. et 02.C 2002). uyJE rm,RcadP vrhd&HlaEckardt Hella & Evershed P. Richard Cramp, J.E. Lucy tal. et 16:0 05 n oa okn eslasmlg tStanwick at assemblage vessel cooking Roman and 2005) plii cd n C and acid) (palmitic or h aietto fadfeec nlcltraditions. local in difference a of manifestation the 1347 > 0) ugsiga cnm here economy an suggesting 40%), 18:0 sercai)aemjrsaturated major are acid) (stearic tal. et

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain

13 13 13 13 Figure 6. Scatter plots showing the δ C values of palmitic acid (C16:0) plotted against the  C(δ C18:0 – δ C16:0)from the archaeological pottery residues, with the ranges obtained from reference animal fats indicated on the left-hand axis. Iron Age δ13C values are previously published in Copley et al. 2005.

1348 oehls ln pctclrwxcmoet r eetdi rnAea ela Roman as or well as availability Age changing Iron in of that detected (Copley but reflection are here residues, residues components the direct argued from wax plant-types a epicuticular is specific plant identify itself it to nonetheless difficult in period, is It necessarily this commodities. of over not selection occurring is transitions mortarium dietary the new indicate either involving Britain practice in cultural recipes. new in or shift apparatus amongst new a or type suggests plants, vessel (e.g. especially This ceramic commodities, non-culinary assemblages. equivalent for Age no or Iron is culinary Roman for there role the and was purposes, unique Age purpose pharmaceutical) Iron this a or whether British fulfilled cosmetic say from cannot they investigated we mortaria, vessels Whilst domestic sites. of other purpose(s) mortarium the the the with in compared whatever components wax that epicuticular indicates plant residues degraded of prevalence striking The Conclusion region non-ruminant the towards (Woodbury compared lie enriched plot to relatively expected the are be of therefore oils would from and signatures fats ruminant isotopic with The oils. plant of indication ln-eie rdcsi bqiosa eryaltestsivsiae ee e ecno be cannot we yet of here, processing investigated the sites on the produce emphasis all to heavy nearly leaves A at Apicius. ubiquitous plant is of and products recipes has oils plant-derived the spices, in this Traditionally, described with as products. meat dishes mixing animal culinary and and grinding plant as interpreted involving been preparation resource of be type to 2006). continued Shaffrey is querns 1987; it Peacock stone that (e.g. Moreover period suggests these. Roman mortaria replace the Roman to in Age in utilised used Iron residues were the fat and mortaria during animal use-wear that produced of form, unlikely were frequency of querns the terms cereals, rotary in grind and comparable to saddle vessel is stone metal that Whilst or Britain finish. ceramic in abrasive no is period there pre-Roman However, the utensils. other from by fulfilled mortaria been that components. previously argued dietary had is novel It of utensils. introduction domestic the reflect the solely, of or necessarily, all not in do reflected therefore is that products Iron (Copley the dairy site in on same observed the frequency the from both the analysed in with vessels observed Age comparable fats residues, dairy pot’ of shift. ‘cooking with prevalence dietary and high well significant unusually mortarium a an compares to is relate vessels there not Stanwick ‘cooking’ does at and Roman Similarly vessels in domestic Age origin Iron from plant The observations components. of unique components of presence of the than frequency rather mortaria, Roman in components these that possible ( vessels is British It with isotope compared former. of residues enrichment the mortarium significant, Xanten statistically of in but contribution slight, displayed the values a for responsible mask be may may oil plant fat animal of concentrations for ocn otiuin srflce ntefua remains. faunal the in reflected as contribution, porcine hltaaye ffua n oaia sebae rmIo g n oa sites Roman and Age Iron from assemblages botanical and faunal of analyses Whilst h otlkl otx o h nrdcino h otru sta trflcsanovel a reflects it that is mortarium the of introduction the for context likely most The but existed, already which function a fulfilling were mortaria that is possibility second A δ 13 C 16:0 , δ 13 C tal et 18:0 uyJE rm,RcadP vrhd&HlaEckardt Hella & Evershed P. Richard Cramp, J.E. Lucy 05 n h ifrnele ntefeunyadaudneof abundance and frequency the in lies difference the and 2005) . and tal et  98 Steele 1998; . 13 aus lhuhti a lob xlie yahigher a by explained be also may this although values) C 1349 tal. et tal. et 00;hwvr iigo iswt high with oils of mixing however, 2010); 05,ipyn ogtr emphasis long-term a implying 2005), p = < 0.002

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain certain that this was to prepare ingredients for consumption; indeed a prevalence of plant- derived components is also a feature of Roman glass unguentarium contents (Ribechini et al. 2008). Neither do the residues support an entirely consistent usage of the mortarium in Britain. The unusual prevalence of dairy fats in the Stanwick mortarium residues is indicative of adaptation to pre-existing localised cultural or economic patterns. We would therefore argue that mortaria may be interpreted not as a direct reflection of a transition to ‘Romanised’ food-ways, but rather as a more complex and selective process of cultural adoption and adaptation in or beyond the kitchen.

Acknowledgements The authors would like to thank the following people for their help and advice in the collection of samples: Norbert Zeiling and Bernd Liesen (Archaeological Park, Xanten), Richard Hobbs and Ralph Jackson (British Museum), David Mason and Lee White (Durham County Council, Archaeology Section), Sara Lunt, Heather Bird and Vicky Crosby (English Heritage), Rob Symmons, David Rudkin and John Manley (Fishbourne Roman Palace), Jenny Proctor (PCA North), Cameron Moffatt and Peter Boyd (Shrewsbury Museums Service), Rob Perrin, Philippa Walton and T. Scott Martin. We would also like to thank Sally Grainger and Kay Hartley for invaluable discussions. We would also like to thank the four anonymous reviewers for their helpful and constructive feedback. Stephanie Dudd is acknowledged for the preparation of Stanwick cooking vessel residues with funding from English Heritage, and Ian D. Bull and Rob Berstan for technical assistance. NERC are thanked for funding the research.

References COPLEY, M.S., R. BERSTAN,S.N.DUDD,V.STRAKER,S. PAYNE &R.P.EVERSHED. 2005. Dairying in ALCOCK, J.P. 2001. Food in Roman Britain. Stroud: antiquity. I. Evidence from absorbed lipid residues Tempus. dating to the British Iron Age. Journal of BLAZQUEZ´ , J.M. 1981. Mosaicos romanos de Cordoba, Archaeological Science 32(4): 485–503. Jaen y Malaga. Madrid: CSIC, Instituto Espanol˜ de DALBY, A. 1998. Cato. On Farming. De Agricultura: a Arqueolog´ıa Rodrigo Caro. modern translation with commentary. Blackawton, CHARTERS, S. 1997. Chemical investigations of Devon: Prospect Books. absorbed lipids and laboratory simulation DICKINSON,B.&K.F.HARTLEY. 1971. The evidence of experiments to interpret archaeological pottery potters’ stamps on Samian ware and on mortaria for vessel contents and use. Unpublished the trading connections of Roman York, in R.M. PhD dissertation, University of Bristol. Butler (ed.) Soldier and civilian in Roman Yorkshire: CHARTERS,S.,R.P.EVERSHED,L.J.GOAD,A.LEYDON, essays to commemorate the nineteenth centenary of the P. W. B LINKHORN &V.DENHAM. 1993. foundation of York: 127–42. Leicester: Leicester Quantification and distribution of lipid in University Press. archaeological ceramics: implications for sampling DUDD, S.N. 1999. Molecular and isotopic potsherds for organic residue analysis and the characterisation of animal fats in archaeological classification of vessel use. Archaeometry 35(2): pottery. Unpublished PhD dissertation, University 211–23. of Bristol. COLOMBINI, M.P., G. GIACHI,F.MODUGNO &E. DUDD,S.N.&R.P.EVERSHED. 1998. Direct RIBECHINI. 2005. Characterisation of organic demonstration of milk as an element of residues in pottery vessels of the Roman age from archaeological economies. Science 282: 1478–81. Antinoe (Egypt). Microchemical Journal 79: 83–90. DUDD, S.N., R.P. EVERSHED & A.M. GIBSON. 1999. COOL, H.E.M. 2004. Some notes on spoons and Evidence for varying patterns of exploitation of mortaria,inB.Croxford,H.Eckardt,J.Meade& animal products in different prehistoric pottery H. Weekes (ed.) TRAC 2003. Proceedings of the traditions based on lipids preserved in surface and Thirteenth Annual Theoretical Roman Archaeology absorbed residues. Journal of Archaeological Science Conference, Leicester 2003: 28–36. Oxford: Oxbow. 26: 1473–82. – 2006. Eating and drinking in Roman Britain. Cambridge: Cambridge University Press.

1350 E E 91 h otra nC atig (ed.) Partridge C. in mortaria, The 1981. – D E H E F 00 esnlvrain nbl ise at cdand acid fatty tissue, bulk in variations Seasonal 2010. – E E E LOWER VERSHED VERSHED VERSHED VERSHED VERSHED VERSHED VANS UNGAIT ARTLEY oni o rts Archaeology. British for Council Don and Hayes Spratt Raymond of north-east honour of in Yorkshire archaeology the in studies monuments: h -at nB (ed.) Vyner B. in north-east, the 415–28. oit o h rmto fRmnStudies. Roman of Promotion the for London: Society 196–9. 2): site series Romano-British monograph and (Britannia Age Iron Late a Green: Archaeology. British for Council London: 39–51. pottery Romano-British (ed.) Detsicas A.P. in mortaria, opudspecific compound nepeaino bobdrsde narchaeological ceramics. in residues absorbed of interpretation B ise at cdadmonosaccharide and acid fatty tissue, rnatoso h oa oit fLno B London 19–31. of Society Royal the of Transactions prehistory. from signals anthropogenic M E ifrn rzn managements grazing different under communities plant grassland mesotrophic M ouet Praehistorica vessels. Documenta pottery archaeological and of fats extracts reference lipid for results of assessments acids: flae rmamstohcgasadplant grassland community. mesotrophic a from leaves of monosaccharide iewo rgnfrpthfo h ayRose. Mary the from Nature pitch for origin wood Pine rceig fteBiihAcademy British the of Proceedings application. and methods interpretation new analysis, residues: of food of survival The 1992. 1:1339–42. 115: Spectrometry. Chromatography-Mass Gas and Chromatography Gas Temperature origin High archaeological by of residues organic of Analysis okn yAiisfruei h td n the and study the in kitchen. use of for Art Apicius The by of Cooking translation critical A book. cookery VERSHED ERGEN .19.LtrIo g n ntv’pteyin pottery ‘native’ and Age Iron Later 1995. J. , UKHERJEE OTTRAM ,B.&E.R .17.Temreigaddsrbto of distribution and marketing The 1973. K. , ... .D G. J.A.J., , .. ..D S.N. R.P., , D S.N. R.P., , H C. R.P., , H C. R.P., , ,R.P,K.J the to approaches Experimental 2008. R.P. , CArsac eot11:4–8 York: 46–68. 101): report research (CBA 1(01:528–30. 314(6011): ..B H.A. , odn Harrap. London: ol Archaeology World 08 neseicvraini bulk in variation Interspecific 2008. . ..S A.W. , 02 dnicto faia fats animal of Identification 2002. . Phytochemistry OSENBAUM ERMAN LAND δ ERON ERON 13 OCHERTY TOTT δ Cvaluesofleavesfrom UDD UDD 13 99 iisa aresof carriers as Lipids 1999. . auso niiulfatty individual of values C CArsac eot10): report research (CBA ,S.C uyJE rm,RcadP vrhd&HlaEckardt Hella & Evershed P. Richard Cramp, J.E. Lucy &G.E &L.J.G ,A.R ,M.S.C ,S.C 9 73–96. 29: 1958. . ,V.S 9 2041–2051. 69: 01:26–47. 40(1): HARTERS AVEN HARTERS GLINTON urn eerhinto research Current Phytochemistry Moorland TRAKER OAD OPLEY 7 187–208. 77: h Roman The ,P.F. 1990. . δ &L.J.G ,H.R. Philosophical 13 Skeleton A.J. & 1985. . VAN &R.P. Cvalues Analyst 354: 71: OAD via . 1351 H R. & Connor A. in stamps, mortarium 1999.The – J. & Wardle A. Neal, D. in mortaria, The 1990. – (ed.) Niblett R. in mortaria, The 1985. – O P M P R P P H ARMINTER HELPS EACOCK EACOCK EGERT SWALD ERON ERON AY rhelgclServices. Archaeological Leicester of University Leicester: 109–110. and 1991 1980 Excavations Leicester. Lane, Causeway (ed.) Buckley England. for Commission Monuments & Buildings Historic 191–5. London: 14): report archaeological Heritage Albans (English St Gorhambury, at settlement medieval (ed.) Hunn for Archaeology. Council British London: 92–93. 57): report Camulodunum research at site industrial Roman early Oxbow. North Lincolnshire in settlement Romano-British and Age niure Journal Antiquaries Sussex. West Lodsworth, at production rhelgclScience Archaeological potsherds. residues buried organic with on associated lipids soil of migration of xod Oxbow. Oxford: North in Lincolnshire settlement Romano-British and Age Iron (ed.) May J. niura Society Antiquarian Cheshire and Lancashire the of Transactions hi eeomn rmA 0t 400. Journal to Antiquaries 50 AD from development their n h oa cnm:a nrdcoyguide Longman. introductory London: an economy: Roman the and 549–69. contexts. various environmental from samples archaeological and of tests analysis ageing accelerated time: through beeswax D eis16:494.Ofr:BiihArchaeological Reports. British Oxford: 429–47. 196): series to techniques scientific archaeology of application the on a conference of Proceedings 1987. (ed.) Glasgow Tate archaeology, J.O. & Slater E.A. amphoras, in Roman from materials resinous natural .1996. J. , ECAVALLAS ..12.Teclnr s fmortaria. of use culinary The 1923. J.J. , ,C.&A.M.P ,C.,R.P.E ,M.,S.C .14.Temrai fMriuu and Margidunum of mortaria The 1943. F. , ...&DF W D.F. & D.P.S. , quern Roman and Age Iron 1987. D.P.S. , Lietracaooia oorps5): monographs archaeological (Leicester ,Y.&K.H rgny eoto xaain ta Iron an at excavations on report Dragonby: BiihAcaooia eot British Reports Archaeological (British xaain fa rnAe oa and Roman Age, Iron an of Excavations Obwmngah6) Oxford: 61). monograph (Oxbow Obwmngah6) 567–73. 61): monograph (Oxbow 01 hmclatrto n s of use and alteration Chemical 2001. . rgny eoto xaain tan at excavations on report Dragonby: OLINART oa n eivlocpto in occupation medieval and Roman VERSHED OLLARD ARTLEY 7 61–85. 67: 2 45–63. 22: 9 1–15. 39: ,L.D 8 641–59. 18: ILLIAMS &L.J.G Archaeometry 98 h nlssof analysis The 1988. . 96 h otra in mortaria, The 1996. . EGRAND 1986. . cec and Science OAD ora of Journal &O. 91 Effects 1991. . hee:an Sheepen: 43(4): Amphorae The The (CBA .

Research What was a mortarium used for? Organic residues and cultural change in Iron Age and Roman Britain

RIBECHINI,E.,F.MODUGNO,M.P.COLOMBINI &R.P. STERN, B., C.D. LAMPERT MOORE,C.HERON & A.M. EVERSHED. 2008. Gas chromatographic and mass POLLARD. 2008. Bulk stable light isotopic ratios in spectrometric investigations from Roman glass recent and archaeological resins : towards detecting unguentaria. Journal of Chromatography A 1183: the transport of resins in antiquity. Archaeometry 158–69. 50(2): 351–70. RUSH, P. 1997. Symbols, pottery and trade, in K. TYERS, P. 1996. Roman pottery in Britain. London: Meadows, C. Lemke & J. Heron (ed.) TRAC 96. Routledge. Proceedings of the Sixth Annual Theoretical Roman WOODBURY, S.E., R.P. EVERSHED & J.B. ROSSELL. Archaeology Conference: 55–64. Oxford: Oxbow. 1998. Purity assessments of major vegetable oils SERPICO,M.R.&WHITE. 2000. Oil, fat and wax, in based on δ13C values of individual fatty acids. P.T. Nicholson & I. Shaw (ed.) Ancient Egyptian Journal of the American Oil Chemists Society 75(3): materials and technology: 390–429. Cambridge: 371–9. Cambridge University Press. YOUNG, C.J. 1973. The pottery industry of the Oxford SHAFFREY, R. 2006. Grinding and milling. A study of region, in A.P. Detsicas (ed.) Current research into Romano-British rotary querns and millstones made Romano-British pottery (CBA research report 10): from Old Red Sandstone. Oxford: Archaeopress. 105–112. London: Council for British Archaeology. SIMONEIT, B.R.T. 1986. Cyclic terpenoids of the – 1977. The Roman pottery industry of the Oxfordshire geosphere, in R.B. Johns (ed.) Biological markers in region (British Archaeological Reports British series the sedimentary record: 43–99. Amsterdam: Elsevier. 43). Oxford: British Archaeological Reports. STEELE, V.J., B. STERN & A.W. STOTT. 2010. Olive oil ZAVARIN,E.&K.SNAJBERK. 1980. Oleoresins of or lard? Distinguishing plant oils from animal fats Pinyons. Journal of Agricultural and Food Chemistry in the archaeological record of the eastern 28(4): 829–34. Mediterranean using gas chromatography/ combustion/isotope ratio mass spectrometry. Rapid Communications in Mass Spectrometry 24: 3478–84.

1352