Pressures on osteological collections : the importance of damage limitation.

Durham Research Online Deposited in DRO: 09 October 2009 Version of attached file: Published Version Peer-review status of attached file: Peer-reviewed Citation for published item: Caffell, A. C. and Roberts, C. A. and Janaway, R. C. and Wilson, A. S. (2001) ’Pressures on osteological collections : the importance of damage limitation.’, in Human remains conservation, retrieval and analysis : Proceedings of a conference held in Williamsburg, VA, Nov 7-11th 1999. Oxford: Archaeopress, pp. 187-197. Further information on publisher’s website: http://www.archaeopress.com/searchBar.asp?QuickSearch=Human+remains Publisher’s copyright statement: Additional information: Use policy The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-profit purposes provided that: • a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult the full DRO policy for further details. Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom Tel : +44 (0)191 334 3042 — Fax : +44 (0)191 334 2971 http://dro.dur.ac.uk PRESSURES ON osn:OLOGICAL COLLECTIONS. THE IMPORTANCE OF DAMAGE LIMITATION A_ A.wa C. Catrcl1, CUrlottt A. Roberti, Robert C. J... way . .d "-drew S. wu.o. Physical anthropology Juu bun taught I,. ｾｨｴ Department 0/Ar'ClIiJLologica/ Sc/eru:es GIllie Un;venity ofBrtldford at boIlt flrtdugradllQk and ptnlgradwta Itw!ls s;nct tile lak 19701. 17Iu stvdy 4JSf!SSQ Me imptJcl of we 011 tM .m-.logicDlly tUrlwd sk<1<uU coll«tlmu by 1M [)q»rtIfttettL A IIU'\Ir)' of G SiJlffpie of lIdull sU/ewn.r jrowI two differmt arcNuologiaJ/ sita, and 1M Mdma of t:laMagr to tlwu shktmu in 1M 10I'7ft of boM elDMIII lou. ｾ fractu.ra aN! swj"at:e erosion oW'r a nlUPlber of years of lISe is rqxH"kd. The rQIIlts s!towed lJUJl 1M Mall)' we sale/orIS SIIffered mast domage tmd loss. orad 1umdJ. lUI aNi I«tlt wtr't' lost ,"ost. Repaint/ and failed rqxU' brrah also OCC1Irnd. ptUtic:wJDrly in ｾ 1Jeayy lISe grotIp. Mild of 1M elemntl gain was Gltrlbflled 10 poor initiDJ recording. Padllging was aLso poor ill grMTdL nw stwfy Iuu ruw1tM in a rr"J'I"Gwl of/uutdJiJtg tutd p«kogiJtg proc<dwos. nee'" """'W lat:roelKt:&M ·It is geuenJly ICCCpl<d IIlalthe: quality or the: information that can be pined &om arebacologicaJ remains is inversely proportiooaI lO the: degree or degndatioo during burial. It is olleo forJoaen, Ihougb, 1halthe: value or lhc:se r=aios is also offecl<d by the: way in wbicb they have bcco _lOd since the rDOmCIlit of dixovay... Just u harmful can be the biddcu damo&< (caused) by poe>< pocbsio.. bod bandHog procticesand low pod< _ . (Sprigp, 1989: 39) Pbysicallbioloaical 0Dlhi0Jl0l0lY bas bcco tough! in the: ｾ or Archaeological Scieoccs .. the: Univemty or Brodford .. bo<b ｾ and postpduotcs I..el. .ince the: IaIe 19101. IoitiaJly claucs wm: devolOd mainly to the ' .....hina of pa1acopaIbology, reflecting the interests of Dr. Keith M.nc.......ct iDstiplCd tnt:hiog in this ..... 1Iowncr. a wider nnge or lUbjects is DOW ｴ｡ｵｧ｢ｾ including ttlOduIes in bosic 0S100I0BY. b..... evolurioo, roreosic -OPOIocY and paIaeopaIhoIosy. These councs r.... part orthe foUowina de.....: BSc ArcbaooIocY. BSc Arcbacological ScicD<ca, BSc Bioan:baeoIo&Y. MA Sci...tilic MedJocb in ArcbaooIocY. MSc in 0sIc01ocY. ｙｳｯＱｾｬ｡ｐ and Funerary ArchaeoIosY. and MSc in Forensic ADtbIopology. The telChing ethos oftbe modules U1iliza the Iabonlory u the: main vehicle. This allo.... "''''mu lO Ieonl aboul the: b..... lkel...., and its voriaIioa. stelelal c:oIlec:Ooaa have bcco acquired lO fulfill Ibis ooed (eapoci,11y .... the: Iut .... ,..on) lO 'CCOcGi!l(lde'e 1be iDaeue ia Ibwtmr aumben. This pipet, tbeeeftn, aimI to IDeSI the iqJed of the prasura of usc mainly by ........... the an:boooioIi<aIIy derived Ilu:Ielal ..11_ _ by the ｾ or Arcbaoo\oaicaJ Scieacea .. the Uaivemty or BndfonI, fslalaod M.m -"ically. the lint baIf or the: popcr IrOCCO the: deveJosw- or pbyoicaI _ _1ocY _bini in its widcat _ in dle U.K. and .. BndfonI. the: bi*-Y. and incn:uina ｾ or the: cwaboa or .... .elelal ..11ee..... and dle types or ..... wbicb Ilu:Ielal moteriaI .mo is put. The second half of the paper focuses on a recent survey of a selected sample of skeletons &om two different uchaeologjcal sites C\D"I.lcd at Bradford, and the evidence of damage to those skeletons in the fonn of bone clement loss, postmortem hctures and surface erosion over a . Dumber of years of use. Refc:rcnce is also made to stonge: cooditi... and standards or pacbging Ilu:Iet.aI malcriaI. The paper finally discusses limilati... or the: S1Udy and ..... =ommeodatioos, with rerer<nce lO o<bcr publisbed wort. aJtbough 0D0lhc:r paper in this volume coven this latter upcct more fully (J.......y et 01.). To the: autbon· ｫｮｯｷｬ｣､ｾ a study such as this has never heeD carried out lOdote. .r PIIys1ca1 Aa"ropolocy: no E"PO_ .. "'U.K. Unlike Nonh America (A.A.A.. I99S). in the: U.K. the: teaching of physical autluopology in ｾｩｴＱ｡ｶｮｕ if it is ｾ｢ｧｵ｡ｴ ..... with an:bacolocY departmc:uts and also a small group or anthropolocY departmeots. There an: approxi..... ly 30 arcbacology depanments in the: U.K., and within that group then: U'C about ten who do auy serio", _bing in physical anthropolocY. maioIy .. the: wwlc:rgraduate level. Those that teach in the area alto ante c:ollcctioos of skeletons for tc8cbiog. IDd councs and modules taught in physical anthropolocY an: usually ovenubscnDcd; for cumple, at Bradford in the IC'DC"SIC'r tUrlIlina from Seplember 1999 lO JanllOl)' 2000). 60 studmu tool: an uocla]vaduale moduIc in H..... 0slc0an:bac01ocY wbich involves Iabonlory _bing. ｾ｟ｔ Although pbysical anthropology i. oot tough! u widely in the U.K.. u in North America, when it is, it is cXbemdy popular with students. Recent media t:OVCf&IC of human skeletal ana1ysis from arcbaeological .ites (such u oM... the: ............ a BBC prognm wbicb regularly .... 3-4 millioa viewe:n) bas comributcd to an iDcn::ased iDta'es:L Disreprdios the: media bype tbere i. obviously a ooed for this teachiDg that requires careful plarmiDg aod tbou&bL ADotber. perhaps more positive. difference btt'ltcco _bing physical anthropology in Nonh America and the: U.K. is lila! in the: U.K. most procticaJly hued -bing is undertaken usinlllu:l..... from arcbaeologica1 sites. This !MY be beca".. tbere bas oot bcco the: pressure or repolriaIioo andI.. reburial (Rose et 01. 1996). and !bus Ilu:Iet.aI oollcctioos have beco ouily available. The public .. Iar1e. and arebaoologists man: spccilically. an: becoming nu:h more aware of the impliatiooJ of this issue in the: U.K (parlrer P...... 1995). However. a re<eDl S1Udy (Orgill 1999) suggests 1hal dle British public do oot appear lO be as concemod as implied in """" or the archaeological literature. Guidelines for the trelt:ment of humID remains in archaeological contexts have abo been drawn up in IreIaod (BucIdey et 01. 1999) and Scodaod on-;c 181 1991) emphasizing the edljcaJ In addition, publicatiool have abo Sc<cI,rv1 consicletotioos HUmlln Remllins: ConseTWltoin. Retrieval and Araalysu appeared regarding burial archaeology and the law in En81and, Wales and Scotland (Gam:tt-Frost. 1992) and the excavation of hwnan remains and their treatment (McKinley & Roberts 1993)_ While the use of anatomical replicas in the fonn of plastic models of appropriate bones and whole sk.eletons may be pertinent for basic anatomy classcs, experience at Bradford suggests that most plastic replicas on the market do not show the detailed anatomical fealW'es of the sk.eleton. and do not mustnlte the larger variation (from obviow> to very subtle) that exists within and between populations. It is considered, therefore, that the teaching of physical anthropology and all its subdisciplines is better served by using archaeologically derived human skeletal material in most cases. may he because the archaeological unit bas storage restrictions and/or they wish the material to be used for the benefit of others. The material is then used for teaching and resean::b. and usually an agreement is drawn up between the archaeological unit and the University. In 1984, the Calvin Wells Laboratory was established following the donation oftbe late Calvin Wells' archive by his widow. Calvin Wells was a Norfolk based gmeraI practitioner who bad an interest in human remains from archaeological sites (especially with respect to palaeopatbology) and. to date, his publications far outnwnber any other physical anthropologist's in the U.I<... (Hart 1983). The Laboratory rapidly hecame a =ognized center for the study of physical anthropology and during the 19805 the numbers of students taught by K.eith Manchester increased, but not substantially. Following the appointment of Charlotte Roberts to a lecturer post in 1989 a new Masters course in Osteology. Palaeopatbology and Funerary Archaeology was established and nm jointly between Bradford and the University of Sheffield. Focus oa Bradford: Teacbt.c of PbysicaJ Aatbropoktgy over the lut Tea Yean The leaching of physical anthropology at Bradford started in the late 19705 with the arrival of Keith Manchester (a general practitionerlprirnary health care physician) as an occasional lecturer in the Department of Archaeological Sciences. At the time, only small groups of students (about 1I}-12) won: taught (mainly palcopathology) in each one of the tenD'S eight weeks for rwo hours. The students used two skeletal collectiOllS acquir<d hy the Department. one through a srudent who bad excavated the site (Raunds, NortbamptonshH<) and the other through worlt Manchester bad done on a collection from Eccles, in Kent. Although initially a course with seven students, it quickly gained in popularity with • concomitant increase in students (1989-2, 199G-7, 1991-12, 1992-9, 1993-11, 1994-18_ 1995-20, 1996-19, 1997-21, 1998-22, and 199914). Runnin8 pantle! with this incroasc in postgraduate students on the course was an increase in lmdergtwiuate nurnhen from 26 in 1989 to 72 in 1999 (199G-34, 199144_ 1992-45, 1993-52, 1994-54, 1995-67, 1996-88_ 1997-90_ 1998-89), reflecting CUl'T'eDl government policies on higher education. As the ovcnll nmnbers of students increased, the numbers of undergraduates opting to take the HUIIWl Osteoarcbaeology module also rose (24 in 1992, 12 in 1993,12 in 1994,24 in 1995,20 in 1996,26 in 1997, 33 in 1998, and 53 in 1999). In additie.-t, students from the MSc in Forensic Anthropology and the MA in Scientific 10 1983, with the arrival of Charlotte Roberts, the number of contracts involving skeletons from archaeological sites increased. as did the skeletal collections in the Department. In the U.K. there bas been a terldency f", the archaeological organizarionlunit, which has excavated the skeletal mat.erial. 10 allow curation 10 be carried out by the institution where the analytical work is carried out. This Total Student Numhe.. U_klng PhyslcaI Anlllropology CoulMSllloduleo (1111I-1_) 00 10 1Ie 00 • 50 "• .. '" ｾ .• 3D ci Z 20 , .............. - 1-.eo 1118O-111Stlil1-82118201311lB-14 11MM-151il85-181f186.871187. . 1. . . . 1818-OO Yoo, Fi8. I - Nurnhen of students itt physical anthropology (1989-1999) \88 Cal/ell el al: Pressures on OSlmlogical Collections Methods in Archaeology can opt to take this module. In 1997 another new muten course (MSc Forensic AoIhropology) wu",-, taking six studeots in 1997, 10 in 1998 and six for 1999. In 1997, new Wldergraduate modules in Biological Anthropology also started (numbers last year were: 67), and in 1996 a module was created in For<IlSic Anlhropology for students taking the BSc in CbemistJy with Pbarmaceutical and Forensic Science (numbers were 36 in the 1999·2000 academic year). Thus. the numbers of hours physical anthropology laboratory classes in a wide range of areas are taught have risen over the yean &om about 16 to about 200 hours over the 24-week teaching year. 1be nmnbers of students taking ｾ based classes in physical anlhropology have also risen (fig. I), and from 1996-7 the nwnbers have ocarly doubled. In odditioo. short laboratory based professional COUI"SC5 accepting about 30 students were nul for one '" two ....b in the summen of 1988 (I), 1994 (I), 1996 (I), 1997 (I) and 1999 (2). A n\Dllbe,- of adult education classes (evening classes, day and ｾ ｫ ･ ｮ ､ schools) have also been organiud during the 19805 and J99Os. all with varying amounts of time spent in the laboratory handIiog skeletal mII<riaI. Finally, the n\Dllbe,of IIIl<Icriroduat and postgnldtwe students (mastcn and PhD) utilizing the skeletal collections for their dissertations bas also iDcreascd over the years. Students ace. to a certain extent. allowed unsupervised access to selected "practice" skeletal material to help with their studies. For the last teo years, students in all of the courses have been instJuctod in the fragile and non-_ble nature nf the skeletal material used in ｴｨ･ｾＮ Benches are covered with prolCCtive mII<riaI (bubblewnp) to prevent damage, and a to place a skeleton into a box diagram iUusaming the ｹｾ is available. 1beIe instructions arc also given to any the collections (Janaway et visiting rescan:bcn who ｾ al. this volwoe). ....... 0 · d Ea: Kmt ... 1910 Ra Sa Hertfordshirt Cbjc..... ｾ Kia GkJuo a mbirt Addin Well Yorbhirt St. Q; North YOfbhirt BiK . ｾ Cbic ｾ RO'llt Pmldiiid r 1982 1987 19U 1911 1990 1991 1992 1991 1991 Hu 1999 1. - Skek:tal colloctiou in the J:lqMbneut of An:baeoIop::aI ScieDc:a, UDivenity ofBrMford. Table well as skeletons that are temporarily cumcd for the purposes of analysis. Due to its large size, the collection is CUlTCDtly housed in four different locations, mainly within the Department of Archaeological Sciences. Over the years, the collection bas been housed in a Dumber of buildings within the confines oftbe University, close to the University, and in areas outside of the main city, depending on availability. None of the storage areas bas, or bad. environmental control. They do not, therefore, meet CUJTCDt Musewns and Galleries ｮｾｳｩｭｯｃ (MGC) standards, and the enviJomnental conditions have varied from dry '" very damp. Inevitably, lmlsport to and from these various locations bas affected the collections' condition. Thus, bwial environment. exeavatioo tocluUq.... 1JI'OCC'Sing, packing, transport, curotion and handling have an Wldoubtedly contnbuted to the current condition of the material. Furthermore, the collection bas never bad a full time curator. This deficiency bu mainly been because of a lack of funds to support such a position, a1lhoogb the department bas been made awar< of the need. I M 21 C1I:remcmt (ok! AtcbIeokJgic::al Scieaca Depc1mau) Wardley H_ (tanpenry ｾ Dell·mall) Scienca - 1912-1913 198}.1984 ... Wardlouae at Wroec., ｾ ｂ , SakI Mill, BtadIord 1985-I986 Block..... (Hor\(Xl A. Curruu ｾ ScieD::a 1986<.1994' DqaibNtit) IUc:bmoad BuildiDa Rncmcm (maiD Uniwnity bWIdiDal ......,. ....... _ _ D.""""""") ｾｉＢ ..- '- Table 2: Storage history for Raunds. Source: Keith (pm.eomm.) Ｎ Ｂｾｍ Materiall ud MedlodJ TMsitD_ The two sites used to survey the collectioo came from cemeteries in Sussex and Nortbamptonsbire. Three hundred and fifty-one skeletons were excavated from the later Medieval (12· -16· cent AD) lepcrl' hospital ofSt. James and St. Mary Magdalc:oe in Chichester, West Sussex beJwecn 1986 and 1987 (Magilton & Lee 1989). A further forty-four were excavated in 1993. Since their arrival in the Department of Atcbaeological Scienc<:$ at the Univemty of Brodfonl, they have been sIDled in the Department's main SUft, and in the I<OCbing and .......b laboratories. The skeletons excavated in ＱＹｾＷ have been in the Department for about eleven yean and, for the ptDpOSCS of Ibis study, wen: named the "Old Cbicbcsta" sample. The: more rccc:ntJ:y excavated skeletons have been in the Department for about a yell and, for Ibis study, wen: named the "New Cbicbeslcr" sample. The other ,;te, Rounds in Nortlwnptonsbite produced three bundted and sixty·thrce skeletobS, which were excavated between 1977 and 19&4 (Boddington 1996). Since 1982 mII<riaI from tbe site has been stored in at least seven locations (Table 2). The tint location, the former An:bacological Scienc<:$ In order to _b physical anIhropology as a Iabontory based discipline, the Deporuncnt bas required skeletal eoUcctiona. The Calvin Wells Labontory ....teI around 1500 skeietoos of vvying aae. from a rouge of g<ognpbic area i:D the U.K. 1beIe lkeletonl were acquired at various times durina the Labon"""s Itistory (Table I). AtIditiouIIy, 1here .... aome smaller and/or pattiaI sites, as 189 ｈｾＱｲｕｊｬ RelM;n.s: Corunwto;f1, Retrieval alld Analysis Department in 21 Claremont. was damaged in 1983 by an explosion in an adjacent garage, and part of the collection was lost since it was stored in the basement of this building. Undergraduate dissenations of the period refer to this 'unfortwlate cxplosion' (Cameron 1984: 13; Thompson 19S5) and, in fact, part of !he work undertakeo by Cameron 'involved separating the skeletons into individual bodies and helping to rebox them'. This explosion probably caused the most damage (in the fonn of fragmentation, and loss and mixing of elements) during the curatorial history of this collection. The bone elements from both sites were marked with Indian ink. although the consistency of elements marked varied. Marking. in theory. sbould etlSUJ't: that the right bones stay in their respective boxes or are placed back into the right box. of the: total c:emetcry, and SI.W. of the adults in - Forty adult skeletons were selected from the two sileS., most of which have been used extensively for teaching and research over the years. Reconstructing the "use history" of the skeletal material, involved establishing areas and pressures of use including: dissertationllbesis work Ｈｵｮ､｣ｲｧ ｡､ｾ mast= aod PhD). laboralO')' elasses, course 'ssessments (e.g. the skeletal reportS by students doinS !he MSc in Osteology. Palaeopathology aod Funerary Archaeology). visiting researcben. summer schools and short courses. The Jack of record keeping. and changes to both course and module strUeturr: (and to delivery within those courses and modules). make establishing exact use impossible. However, it was believed that the resulting figures would give an idea of the minimum usc of the: collections, and which skeletons were used most frequently. Of course, this is only based upon Icnown use; it is possible that "less frequently" used skeletons have aetually been more frequently used than believed. Thirty skeletons wen: selected, on this basis, from Chichester (including ten "New Chichester'" to represent a "light usc"'noo-usc group, and twenty ""Old Chichester" to represent • ""beavy use" group). Ten skelClOOS &om Rauods, also r<prcsenliog high aod light usc groups. were selected. Of the "New Cbicbc:s'Cr'" skeletons selected (Table 3), five bad been used fOf teaching for the first time in 1998-9 while the rest bad never been used except by a visiting researcher in July/Augu.U 1999. It was appomtt that noo-adolt sIc.l..... from both ,ites were used far less frequently than adult skcletoos. Due to this, and to the diffcrenc:es in aize and fragility between adult aod noo-adolt skel...... it .... decided to conc:cntrate 00 adult skeletons. The"leu frequently" used sIc.l..... teoded 10 be Iesa ,..,11 preserved, poasibly !he ...... wby !bey ...... not used as often.. Although the Chichester skeletons have been in tbe Department for a sborter time (II as opposed to 17 yean). the skeletons have been used in roughly the ume number of undcrgroduaIe aod postgrad_ (Maslen) dissenations as the Raunds material. Olicbesttz is also UICd more frequently fa< !he skeletal ropon ............ by !be MSc aod by 1aboralOty eiaslea becaute of !he high -u. proportioo of pothological examples. For exampl•• 124 OUcbester tkeletonl were used in various clutes (37.6"'. . .... ""'" ........ .... """ ca OMCNee .... 0" ｾ C40 __pIes ｾＢｯｬ the group). wbil. only 13.7% (51) of !he IOtaI Raunds sample v.'U used comprising 26.4Y. of the total adutts for this population. Oven.1I the Chichester material bas been used more intensively than the RauDds material. C79 CIU el21 Cl42 020 02. 07' C'4I CUI CI23 CIS! CII' 0272 0273 C," .a II C317 C339 C,.. C". 0" 04' 0'" C353 C357 C36' - .... n-y IJ%I 1U02. R504. u. ""'" ""66 R5202 "'207 RS323 ""64 "'22' "'287 02" eJO' cm • • 1 , Table 3 - Skelctoos selected for study M_osn A standard recording fonn developed specificaJly for lhc study was used (Appendix I). Particolar fca..... wert noted. lAss ofanullb Two indicators of damage were recorded by condition scoring: loss of skeletal and dental elements and physical damage to the elements.. To ISIeSS the loss of elements, the elements prescot at the time of this study wert c:ompan:d with those proscnt 00 !he original =ording form. If an clement wu present and matched Ibc description given in the original recording form. it was given. score of ODe. If it was present on the original fonn but now absent it scored ZCfO. Where only partial elements survived in comparison 10 !he original des<:riptions lhese were scored u fr.ctioos. Tbc: cum:nt CODditioD score 9Io"'lS sublrw:1<d &om !he original rcconIlO ...... !he number of elements lost. A DCptivc TC:SUIt indicated a lou and , positive result • pin. The original Chicbcster recording fOl'lDl give the number of fragments present but. unfortunately. the Rauods Corms did DOL The skeleta.l elemcn1S were divided into larJe elements (•.g. 1_ boaes of arms aod legs). baDd demenu aod fool .Iemeots, plus teetb, in order 10 ...... Ioas 10 particoIar pans of !be skel...... In !he cue of !he . - - aod ribs. loss assessment proved quite difficult for a aumber of reasons but IpKC prevents dilCUlSion. The proportion of lott elcmmts compared 10 the: total 1011. elements was also caJco1aled. 11 .... expected tbat IDOr< small.lemeots (,",' u bands aod feet) would be lost thaD 1arJle elements. C..1idD- _,_ _ The OOIIdition of !he . _ .... compared to lb< descriplioos 00 !he rcconIing form in aD alt<mpl1O ...... !he damage austained. I'boCDgnpbs of oome of lb< elements were: abo available IDd were uted as I 190 Coffell et a/: Pressures on OsteologiCQl Collections comparative tool. It 'NU assmned that the photographs were taken soon after the skeletons enten:d the collection, but this could DOt be proved, since there were DO dates linked to the photographs. elements in the: base of the box (i.e. those with elements above but 00 elements below). The presence of postmOrtem damage was rcc:ordcd for each skeleton. Dunage that 0CCWTCd in the burial environment was differentiated from damage that bad occ:um:d since excavation. Tbe following criteria were used to identify damage that occurmt in the burial environment: • staining (Le. similarity in color between the surface of breaks and the cortical surface), • angularity (i.e. by the more rounded edges to the: break, • • presence of soil in the break surface, and presence of soil in exposed trabecular bone. FiS 2 _ Lou of periosteal (pathological) new bone formation from tibia Repaired breaks and failed repain. where either an adhesive bad failed or the element bad been re-brolcen, were also noted. Results ｌｄｓＤ Ｏｾｄｕｓ OveralL. 72.5% of the: fony skeletons assessed bad lost elements, or parts of elements or pathological lesions (fig. 2). Of this group 4()-1. had lost large elements (or parts of large elements), 42.5% bad lost band elements. 40"1. bad lost fOOl elements and 32.5% bad lost some teeth (Table "). Wbeo the data was divided into different use groups it became apparent that more of the ""heavy use" skeletons bod lost elements (94.1 %) tbao !be "light usc" group (76.9'1.) or the "New Chichester'" group (30-1.). Similar trends are apparent in the element subgroups (Table 4 and fig. 3). ,.dll,ilt, pon As of the study !be staodanI, adequacy, suitability, and types of packaging materials used were also assessed.. The order in which elements ｾ pecked was eumined by dividing the box into 1a)'a'S and scoring elements according to !be layer in wbicb they wen: found. A scon: of ODe was applied to elements, which lay on top of others (i.e. those with DO elements above), a score of two to elements sandwiched vertically (i.e. those with elements above and below), and a score of three was applied to -------------- ....... 1O.00'Il. r .lMge ..."'" I-------.Foo< DTeeth_ ..."'" e:....,. C:lWC lIM e..t.gcdle .....- .......... ' Fia. 3 - Percentage of Ikcletom with lost elc:mc:nts by use IfOUP or 6.5%) or "New Chichester" (21.5 or 2.5%) - Table 5 and Fig. 4. The latter score was skewed by the loss of 0.-11, 174.6 elementa (4.6%) bod beeo lost from an initiollOlal of 3779 (including t<eIb). More elements wen: lost from !be "heavy use" group (total 62.\ or 4.11'0, excluding t<eIb) <XIIDpII<C! to !be "light usc" group (49.0 21 elements \91 from skeleton C348. Human Remains: Conservaloin, Retrieval and Analysis % of Elements lost by Skeleton •...,. r - - - - - - - - - - - - - - - - - - - - - - - - - - - , •.•....,..,. ,...,. --...- ...... .... a.m.", Group Fig. 4 - Percentage of skeletons with elements lost by element group ｴｨ･ｲｦｯｾＬ were lost from the higher use groups. and this was also in the sub-groups except for the hand elements (affected by the loss ofhand elements from C348). The percentages of elements use" group. which can be explained by the initial number of elements present in each group. The lost tended to be higher in the ｾｉｩｧｨｴ "heavy use" group had an aV'"8e of90 elements per skeleton compared to the ｾｉｩｧｨｴ use" group that bad Aft average of 58.1 ､･ｴ｣ ｬｦｾ -- - - - - - - - - - - - - - - TOIIII Foot E_.S7% (ill Fig. 5 - Chichesttt: heavy use group: total clements lost On the: whole, the proportions of teeth, band and foot elements lost were greater than the proportions of large elements. However, the proportion of large clements lost in the Chichester "'heavy U$C" group was greater than for the hand clements and teeth, and was almost equal to the proportion of foot elements lost (fig. 5). teeth lost from el2S, resulting from the loss of the entire mandible and maxilla, nor the four teeth. lost from C40, because the teeth wc:re initially glued into the wrong "",kels. Canines proved 10 be Ibe most froquently lost teeth, followed by the promol.... AJ expected, mort single rooted teeth wert: lost since they arc easily dislodged from the alveolar bone. Loose teeth were present for 1weD1y-nine skeleIOIlS (72.5% of the sample), ond repmented 191 (31.2'10) of!he initial oumberofteelh. The proportions of different tooth types lost arc shown in Figure 6. This does not take into account the twenty·two 192 Co./fe// et aJ: Pressures on Osteological Co//ections - ----- ----Tooth loas: All Sttes -----Fig. 6 - Tooth loss for both sites by tooth type Frail bnds IUUl S_rfllU tluIqe The number of fresh breaks and surface damage was far greater in the "'Old Chichester" skeletal material. Material from Raunds showed less post eltC8vatioD damage, possibly due to differences in bone preservation between the two sites. The Chichester material, although more complete, is more friable, whereas the Raunds material is more robust despite being more fragmentary. There: may also be • recording bias as fresh surface damage is more easily visible on the Otichestcr material, since it is darker in color. Photography of some of the elements, proved to be the best comparative tool. 35mm color slides were available for six of the selected skeletons., an from the "'Old Chichester' group. Direct comparison of the elements with the slides showed conclusive evidence of recent 1»<aIcs, particularly in the cnmwn of CI87 (fi8_ 7)_ The slides taken of this individual also revealed that reconstruction work had been strerIgthened some time after the slide was taken. ReptIirU bmW IUUI faiJd "fHIin Repaired breaks and failed repair breaks were more conunon in the "'Old Chichester" heavy use material. The incidences of repairs and failed repairs in the Raunds material were almost always restricted to the craniwn. The percentage of failed repairs was low in the "New Chichester" material (around 6%) but was higher in the: older material (around 30%). Analysis suggested that reconstruction of some bones may have caused some of the fresh breaks. Problems encountered with the various types of adhesive used on the collections included: gluing of teeth into incorrect sockets. and failure to clean break surfaces prior to repair, resulting in poor alignment of adhered fuagments contributing to faiturt: of the join. BlutackTH was apparently used to bold teeth in plact:, which is problematic, since this substance is radio-opaque. Traces of Blu-tack1l'I were also identified on other elements. Masking tape: was used to reconstruct elements, or for the pUJP05eS of numbering ribs and vertebrae, and was oftco left in place (fig. 8). This may relate to student use for specific purposes; for example, there: was a correlation between the use of tape on skeletons used for skeletal reporting by MSc in Osteology, Palaeopatbology and Funerary Archaeology students. GtIUtU datefttS Surprisingly, the total percentage of skeletons gammg elements was 62.5% on average. More skelctoos gained hand and fOOl. elements than large elements (Tables 4 and 5). More RaWlds skeletons gained elements than did Chichester skeletons, and the correlatiOtl between elements high. lost and gained ｾ Fig. 7 - B...uge of skull following =onstrueOoo (CI87) 193 Hu.mtJ1I RnruJi1l.S: ConserwJloin. Retrieval alld AltO/pis Material from Raunds tended to be packed in such a way that bones from one side of the body were in one bag, which led to large elements being in contact with smaller bones. All the Raunds and "Old Chichester' skeletal elements wt:i"e muted with Indian ink. The "'New Otichestcr' skeletons were unnwted. Cantlon ofSkdetaJ CoUertio•• at Other IDldtudo.1 A small survey of a nwnber of institutions using human skeletal collections for teaching was carried out to provide a comparison to the curation and handling of skeletal remains at Bradford. The following questions were asked: Fig. 8 - Excessive use of masking tape Packaging All the "Old Chichester" skeletons and most of the Raunds material were packed individually, i.e. one box. per skeleton. Some of the Raunds skeletons share a box. with one or more other skeletons. All the "New Chichester" material was packed in two box.es per skeleton (oot • separate skull box. hut two box.es of the same size), except 031, which was in tbn:e. In ten cases the length of the loogest element (usually the femur) ex.ceeded the length of the box.; in all cases the femur had been wedged into the box diagonally. This is not TeCOnunended practice. Stroud (1989:48) recommends boxes, 'large enough to take a complete post cranial skeleton comfortably'. Fragile elements such as the cranium and maxilla should have been packed in Layer I (the \lppCl ioost layer) with the heavier and more robust elements such as the lower limb bones in Layer 3. This was DOt, however, the case. Three (15%) of the "Old Chichester" had their crania packed in Layer 3 (the lowest layer), one (50/.) had its mandible in Layer 3, and three (150/.) had their mandibles in Layer 2. The maxilla, in one case, was packed in Layer 3. Furtbermorc:, in one instance a large heavy non-buman bone was found in a bag with the cranium and facial bones. There were abo several instances where ribs, and pectoral and pelvic girdles were packed in Layer 3, and there were three cases (15%) where the lower limb bones were packed in Layer 1. The RaUllds skeletons were in bags with all the elements from one side together, while the "New . Chicbester"' skeletons were in more than ODe box. There was a tcDdency for the ribs of some of the Raunds skeletons to be packed at the bottom of the box. in Layer 3 (6 skeletons or t;()-I.). How large are your teaching collections? • How many students are taught physical anthropology? • • How many yean have the collections been used for teaching? How many hours per week are they used? • • Arc they available for visiton? • Arc casts or anatomical specimens used in preference to archaeological material? • Is technical support available for curatorial purposes? • What are the ｳｴｯｮＮｾ conditions and packaging used? • Who prepares and puts away specimens ｾ for laboratory classes? • Is protective padding used to prevent damage? • Do students wear protective clothing? • Is a handling protoCOl in operation? • Is twxll.ing monitored byanybody7 Although the intention is to undertake a mort compf'C:bensive stUdy, the results showed a number of similarities to the data reported from Bradford. The size or skeletal collections varied from less than 100 to over 18,000 individuaJs. The numben of students Iolcing physical anthropology classes ranged from 5-10 to 40-50, with the avenge being taught over any one year being 1S25. The number of yean that courses have beeo run ranged from three to over sixty years. Laboralory classes were nm &om two to 32 boun per week for two to 4{l weeks per year. Tbe most intensive teaching was for 32 hours per week for 40 weeks. although this was very much the exception. The lowest usc was for two bours per week for ten weeks of the year. All the collections wert available to visiton, Cuts of apccimens were prefc:ntld II four places., primarily for demonsIratioo purposes or for nrc examples. Ai only one institution was there any form of access policy. Only two institutioDl bad technical support (at one,' pan-time student). The IarJest collccrioo was the only one, which had • fuJI-time permanent technician. Elements from the "New Chichester' skeletons were mostly unbagged. However, if they were bagged., usually 1hc feet, bands, venebne and ribs benefited. All lbc: skeletons from lbc: "Old Chicbesu:r" group and from Raunds bad initially been bogged, bitt in sevenJ cues elements were loose in the box. All the bagcd "New Cbicbester" material was in sealable plutic bags, as were 11105' of lbc: "Old Chicbest<r" skeletons. All lbc: Rounds material was bagged in non-sealing plutic bags. which were sometimes stapled or closed with a paperclip. Bubble wrap or tissue paper was used in three institulioitS for Fehging; IDd in aootbc:r boDes 1lm'e pKkcd in plastic oc paper hap. At aaothcr ins1it1llioo _ _ pills oapbtbaJmc baJh IS ... insect repellent, 1lm'e UICd. T'IW p1a<es did ... provide &oy special packin8 foc &agile '" pathological specin>ens. Another two kep< ftogile '" important IpccimenJ ICpU'ate from the main collection. "I 194 Caffe// et oJ: Pressures 011 OSleological Collections four institutions a designated person laid out and put away material for talxntories, and in one case the stUdents were responsible. In all cases material was laid out on the day and in most it was packed away the same day. Access to the laboratory was denied in most cases dwing the period between the material being laid out and the class itself Most institutions used protective padding on benches, although one did not and one used protection for fragile . specimens only. In all cases no protective clothing was used (e.g. laboratory coat, gloves, dust masks), although gloves were available at one institution if needed. Instructions in handling and repacking were usually given. although the fonnality varied. Only one institution had a formal access agreement for visitors, and visitors were only monitored at ORe. Overall, there appears to be little standardization of care for human remains at teaching institutions. Unfommately, time constraints meant that only a small sample of the total collections at Bradford could be studied and only a IUnited nwnber of institutions surveyed. However, on the basis of this study a nwnber of recommendations for the future treatment of skeletal material cwatcd by institutions and used for teaching and ｨ｣ｲ｡･ｳｾ can be made, although another paper by Janaway et al. (this vollunc) deals with these in more detail. Discuado. ud rftoau.adatioal Teaching physical anthropology to large nwnbers of students not only puts pressure on staff and on available laboratory space, but also 00 the skeletal collections used. The obvious result of increased student nwnbers is that more students handle the material and thus increase the risk of damage. There is also a greater likelihood of material becoming lost and mixed. ｆｵｲｴ｢･ ｭｯｾＬ increased student numbers often mean that the same laboratory class may be run several times, which increases the exposure of the skeletons used. The less obvious result of increased student numbers is the preuW't: on teaching staff and technicians. Staff bas less time: to prepare laboratory classes and to repack material beflm the next class. The temptation is 10 leave material out for subsequent classes, thus increasing the risk of loss or damage when people use the laboralory in the intervening periods. In _""" with limited time available, material may not be repackaged properly or retu:mod to its com:cc box. Curation is a fulltime job, especially wben ISOO sltele10llS It< used up to 810 boun of per week during !he ocodemic year. Time, money and dedicated staff are required to ensun: best elements were lost most frequently. The problems encountered when assessing use damage of skeletal material highlights the necessity of thorough documentation. It was originally assumed that clement loss would be ｾｬ｡ｴｩｶ･ｬｹ easy to assess. However, due to the inadequate nanae of the original documentation, in some cases even assessing loss proved difficult The higb nwnw of gained elements, particularly for the bands and feet, also suggests inadequate initial recording. In addition, the level of documentation required to assess the incidence of fresh damage to the material was lacking. Although the COnditiOD of each element was compared to the original descriptions, they did not have the degree of accuracy or standardjzation required to assess v.'bether the damage was recent or not. Only in cases where there: were large fresh breaks was comparison with the original forms conclusive. For example, the patbologicalleft tibia of C28 was recorded as being "complete" 00 the original form. but the proximal end is now shattered into three fragments with sevenal more missing. Attempting to establish more minor instances of surface abrasion proved impossible. Bone loss and damage occuned more frequently in the ""heavy use" group of skeletons, supporting the hypothesis that handling causes damage. Elements may be lost as • result of their removal from boxes and of failW'e to return them to the correct box. If elements are not labeled then this results in permaoent loss of material. The presence of loose elements in the bottom of skeleton boxes from Ratmds and ""Old Chichester" suggest that elements removed for study were not re-bagged before returning the material to its boll.. All bones and fragments of bones should be marked with the site code and skeleton number to reduce the risk of loss or mixing of skeletons. However, any labeled elements rctumc:d to iocorTect boxes art effectively lost, as recovering them would require looking through all the boll.es containing skeletons in the collections. Any fresb breaks will also produce WlIabelled fragments or elements, thus increasing the chances of permanent loss of material. Elements may also be lost during unpacking and repacking of skeletons, as small elements may be caught up in packaging or, if dropped on the floor, can be easily overlooked. Loose teeth are .Iso potentially easier to lose, as they are not held in. and protected by, the mandible and maxilla, and those that are replaced in their sockets have a tendency to fallout .gain duriog handling. It is not so easy to overlook. large element, so loss in this case must imply either that the material was returned to the wrong box or that the elements were not returned at .11. Six skeletons in the sample had labeled elements from another skeleton and labeled non-human bone was .lso found. Furthermore, it could be assumed that an increase in the number of fragments represents recent breaks, and • reduction represents loss ofmater1al. However, there are a number of problems with counting and comparing the: number of fragments. Failure to rccon:I elements and fragments, or breakage of skeletal elements., could oecur practice. Over-Il. the results suggest that the more skeletons are the cbaIlce is tbalthey will suffer loss handled !he _ of ell:lIlCllts. A higher pert:CI1ta8" of sltele10llS from the "Ileavy \lie" group suffered loss compU<CI to the light and unused groups. in gcocral, the Dumber of elements lost &om the '"heavy use" group was greater. However, the loss of material from the light usc groups resulted in • 8JQter proportional loss of matA:rial compored to the ioitially beaer preserved "Ileavy ""," slcele1OllS. Smaller 191 Human Remains; Conserwtoin, Retrieval aNi AfIQ/ysis simultaneously, with loss and gain of fragments from recent breaks canceling each other oUL I)jfferences in the identification of fragments, and in opinions of the minimum size for a fragment to be included in a count, will result in differences in numbers. Element gain may be the result of inadequate: original recording whcrt: elements (most cormnonly hand and foot bones) or fragments art: incom:ctJy identified. On correct identification the number of fragments would increase: from the original. Owing lhis study, whole elements W'a"C found in the "'fragments" bag in a bolt. Tbc:se were most commonly tar>aJs. caJ»&1s and phalanges. Idcotifiable ftagments of other bones were also found, indicating inadequate: initial documentation aodIor mixing of elements during use by students. Mixing of material from other skeletons with the skeleton under study increases the number of elements present. but the gained elements were usually labeled with the correct skeleton nLDDber. Hand and foot elements art: small and therefore are possibly more easily ove:rlookcdl misidentified! miscolmted dwing initial recording (depending on the experience: of the observer), elements may be lost from a fOOl and gained by a band. Frequcot misidentification and incorrect side assignation of band and foot elements were DOted for the Ralmds material. There was also DO differentiation of proximal. intennediale and ltistal phalanges and frequently DO identification of right and left hand and foot bones. To complicate: the matter there was DO diagram, or table, of bones present and this information bad to be deduced &om the written comments. The documentation used may reflect the time period (knowledge and research objectives) when the skeletal material was recorded but this also bas implications for the data actually recorded. The reader is rtferTcd to suggestions for initial recording of skeletal material, which are discussed by Janaway el aI. (this volume). evidence, in this study, are that people tend to photograph only the interesting elements (e.g. pathological), and tbat even when an element is photographc:d, the relevant put may nol be visible for a variety of reasons including focus and lighting. For human skeletal analysts to be familiar with changes due to postmoncm damage and use, ongoing wort must define the criteria for rec:ogn.ition of damage, particularly at the macroscopic level. The physical (and bow anlbropologiSl a1so needs to OOIlSKIer _ many) elements of the: skeleton are going to be rodiograpbed and photographed. or even cast; these: forms of data art: essential for the original record of the skeleton and will help with assessmenl of future: usc: damage:. Ultimately, if these problems and rccolilliavlarions ｾ not .aendcd to, the CODdition. and therefore value: for te:aebing and research, of skeletal collections will inevitably decline. As a consequence, the justification for retention of skeletal material for curation and SIUdy will be difficult to support.. Aa.owtallemttltl The autbon wish to thank the foJlowiDg: Arnold Aspinall, Anthea Boylston. Peter Gillab. Mory Lewis. Jason Maher and K.eith Manchester for help with reconstructing the storage hUtory of tbc: s1teleul maleriaJ S1Udied. Doo Ortner for help with distinguishing different types of breaks in skeletal elements and for providing some of the "'before" illustrations, and the secretarial staff in the Department of Archaeological Sciences for infonnation on studenl nwnben. Jason Maher produced Ihe black and while prints of original color slides. This work comprises put of an MSc thesis by Anwen CaffelJ, student in Osteology, Palacopatbology and Funerary Arebaeology. RtftrtD«l American Anlbropological Association (1998) A.A.A. 10 ProgrturU. A Dinc:tory of Memben. Arlington. VA: A.AA 1998·99. Recent breaks tended to occur more in skeletons in the "heavy use" groups especially Ihe "Old Chicbester" skeletons. Post eltcavation surface damage wu quite pronounced in the "'Old Chichester" '"heavy use" skeletons. The usc: of adhesive wu more common in the OUcbeste:r material, and in the bunds material was restricted to crIJli.II =ons1NClion. The &esb breaks TCCOOlod might . not have oc:currod during handling; tome may be due to excavation, post-ucavation processing or initial recordin&This may explain the pre5CDCC of fresh breaks in the ""New Cbicbestcr" group of unused material. However, if aU fresh bn:ab wen: caused in this way • more equal distnbution across usc groups would be expecc.ed. The: number of breaks prncnt in the umuc:d material may be r<pnIcd .. a hueline against ..mcb additiooal damage would inlti.... damage cauaed by handling. III _ fulur< study. pbo<ogropbK: ev;dencc should play a 1arJler and man: imponant part when comparing pre- and _ damage beca"" this prov;dcd inltisputable evidence f... recent breaks in elements &om six of die selected skeletons. However, the cost implications would be considerable. The limitations usiDg pbocograpbie ｇｵｩ､ｾ Boddington. A. (1996) RmmdJ F",melb: T1te A"glrrSaxOIl Church aNi Churchyard. London: English Heritage. Buckley. L. Murphy. E. ｾ O' Donnabbain, B. (1999) 770, Trml1MJtl of HlUrUl" RemailU. Technical Paper for Arebaeologists. Irish Associatioo of Professional Arehaeologists Cameron. A. (191W) n.- No.-Sp«ifk St= lNJk4wn from an A.g/o.Saxoto P0J"'14tioo at RauN1s. Unpublished ｾ dissaution. Uoivenity of Bndfonl. Gumt·Frost. S. (1992) 770e Law and BwrW A",M.oJogy. lostilUle of Field An:baeo\ogisla Tcc:boical Papet 11. Binoinglwn: lostitule ofField An:baeo\ogisla. Hart, G.D.• eel. (19g3) m-ue i. AodDot Man. ToroolO' CWte Irwin. IliR..-i<: Scot1and (1997) no. T . - - of H_ R...,bu bl A",Iuuoiogy. Edinburgh, Histori<: S<otland. Magi'-. J.R. ｾ Lee. F. (1989) The l.cpc< Hcapital ofSt or 196 Caffell el ill: Pressures on Osteological Collections James and St. Mary Magdalene, Chichester. In C.A. Roberts, F. Lee &, Bintliff, J., eds., Burial Archaeology. Current research Methods aM Developments. Oltford: British Archaeological ReportS British Series 211: 249265. McKinley, J. &. Robcr1s, C.A. (1993) Excavation and Post· Institute of Field. excavation Treatment ofHuman ｾｳｮｩ｡ｭ･ｒ Archaeologists Technical Paper 13. Binningham: lnstitute of Field Archaeologists. Orgill, R. (1999) Re/igiollJ Beliefs and Actitudes towards Archaeological Investigations of Human Remains in Contemporary Smain. Unpublished undergraduate dissertation, University of Bradford. Parker Pearson. M. (1995) EtlUcs and the Dead in British Archaeology. 17Je Field Archaeologist 23: 17-18. Powell, F. (1996) The Hwnan Remains. In: Boddingron, A. Raunds Funeelu. 17Je Anglo-Saxon Church and dwrchyard. London: English Heritage. 113-124. Rose, J.C., G=n, T J. & G=n, V. (1996) NAGPRA is Forever: Osteology and the Repatriation of Skeletons. Annual Review ofAntllropology 25: 81-103. Spriggs, J. (1989) On and Off-site Conservation of Bone. In C.A. Robens, F. Lee & Bintliff, J., cds., Burial archaeology. Current Research MeMotU and Developments. Oxford: British Archaeological Reports British Series 211: 39-45. SlrOud, J. (1989) The Processin8 of Human Bone from An:hIeological SileS.lo c.A. Robens, F. Lee & Bintliff, J.• cds., Burio.l arcMeology. CU1'Te1lt Research Methods and ｾＯｯｰｭ･ＢｴｳＮ Oxford: British Archaeological Reports British Series 211, 47-49. Thompson. J. (1985) Vertebral Palaeopathology: a Case Study from Ute Anglo-Saxon Cemetery at Rllllnds. NorthampwltShire. Unpublished undergraduate disserwion, University of Bradford. Blocropbln Anwen Caffell received a BS<: (HODS) in Archaeology at the University of Bradford. in 1991. Her placement year was spent with the Yark Arcbaeoloaical Trust CollSCrV1ltioo Uboralori.. (6 moolhs) and the Museum of London Archaeology Service (6 months). In addition she bas excavated on various archaeological sites and bas eq>erience in excavating inhumations. She completed ber MSc in Osteology, PabeopaIhology and Funerary Artbaco10gy at the Universities of Bradford and Sheffield (1999), funded by the NERC, and is currently • PhD student at the Univer>ity of DurIwn. an MA in Environmental Archaeology and a PhD in physical anthropology, she was formerly a Senior l.ecturtr al Bradford. Her main area of research is focused on the evolution and palaeoepidemiology of disease with a partieular concentration on the infectious diseases. She taught biological anthropology at Bradford betwCC'D 1989 aDd 2000, is co-author of The Archaeology of Disea.re' (1995) and is CW'Tetltly involved in writing books on Health and disease in Britain through rime, and on tuberculosis in antiquity. Robert Janaway began excavating in 1970, throughout the decade he continued to excavate on a range of sites including a number of sites with inhumation burials. He completed his SSC Archaeological Conservation at Univmity College Cardiff. While employed as a conservator at the University of l...ceds, be instigated the conference that led to the edited volume Death. Decoy and ReconslnlCtion (1986). In 1986 he moved to the Department of Archaeological Sciences. Bndford Univenity, where be currently balds a lecturing post. His current research interests include the taphonomy of inhumation graves. Andrew Wilson completed his BS<: (Hans) IS an Archaeological Conservator at the Instiblte of Archaeology, Univmity College London (1994), and then worked for South Eastern Archaeological Services and as a research intern at the Conservation Analytical Laboratory [SCMRE) Smithsonian Instirution, Washington. D.C. He completed bis MSc in Osteology, (1995-96). Palaeopatbology &, Fooerary Archaeology at the Universities of Sheffield & Bradford (1996-1997). He worked as a Contract Conservator for WilBhire County Council Conservation Service (1998). He is cwrently researching a PhD in hair degradation at Bradford University funded by The Wellcome Trust Bioan::haeology AwanI Scheme. Addresses: Anwm Caffell, Department of Archaeology University of Durham South Road DwlwnDHI UK Charlotte RobertsSame as for Anwen Caffell Robert Janaway Department of Archaeological S<:iences University of Bradford Brodfonl. West Yor!tsbin: BD7IDP UK Charlotte Roberts is currently a Reader in Archaeology at Andrew Wilson the Univer>ity of Durbam, U.K. Tnined initially as • &enen.J nune, with an BA (Hons) archaeology fint degree, Same as for Robert Janaway - Author to whom corTeSpoDdence should be addressed 197 View publication stats