anre-2013-0007

8/12/2019 anre-2013-0007

1/27

ANTHROPOLOGICALREVIEW Vol. 76 (1), 2349 (2013)

Differential preservation of childrens bonesand teeth recovered from early medieval

cemeteries: possible inuences for the forensicrecovery of non-adult skeletal remains

Bernadette M Manifold

Department of Archaeology, School of Human and Environmental Science,University of Reading, Whiteknights, Reading, UK

ABSTRACT: The skeletal preservation of 421 non-adult skeletons from four early medieval sites in England,Scotland and Wales were compared to assess whether geographical location and geology have an impact onoverall bone preservation of childrens remains in the burial environment. Skeletons were examined from

the cemeteries of Auldhame in Scotland, Edix Hill and Great Chesterford in England and Llandough inWales. The bone preservation was examined using three preservational indices: Anatomical preservationindex (API), Qualitative preservation index (QBI) and the bone representation index (BRI). A similar pat-tern existed across all the sites with regard to what bones are preserved, bones with relatively high density,such as the temporal bone of the skull, the long bones of the upper and lower limbs tend to be abundant inthe samples, with the more small and fragile bones, such as the facial bones tending to be less well repre-sented either as a result of low bone density or due to loss at excavation. The study of the dental elementsalso revealed a pattern with regard to what is preserved, with high numbers of molars and incisors found.

This may be related to both the size and number of roots; but also the position in the mouth which mayoffer protection against loss. A difference in preservation was observed between the sites and the classesof preservation, particularly local differences between the sites of Edix Hill and Great Chesterford. Fromthis study it remains unclear as to the extent the role of geology has on the non-adult skeleton, but the

results of this study show that age is not the dominating factor in bone preservation as previously thought.

KEYWORDS: bone preservation, deciduous and permanent dentition, under-representation, taphonomy, fo-rensic science, age

Introduction

This paper aims to examine bone preser-vation in a number of non-adult skeletalassemblages from early medieval Britain.

The skeletal remains of infants and chil-dren are often limited in numbers whenrecovered from cemeteries and this hasgiven rise to discussions on the possiblecauses such as taphonomic processes,

Preservation of non-adult skeletal remains

Original Article: Received June 02, 2013;Accepted for publication October 10, 2013DOI: 10.2478/anre20130007 2013 Polish Anthropological Society

Bernadette M Manifold

Brought to you by | University of the Philippines DilimanAuthenticated | 202 92 128 135Download Date | 6/11/14 8:10 AM

8/12/2019 anre-2013-0007

2/27

24 Bernadette M Manifold

burial practices and or excavation tech-niques (Mays 2010; Bello et al. 2006;Henderson 1987; Nawrocki 1999). Pre-vious studies have shown that non-adultremains are less well preserved thanthose of adults (Bello et al. 2006; Djuricet al. 2011; Buckberry 2000; Guy et al.1997). Taphonomic factors can be divid-ed into two forms: intrinsic (resistance ofbone) and extrinsic (environmental inu-ences), both of which exert inuence onthe long term survival of non-adult bone.The most prominent intrinsic factor isthat of age (Manifold 2010; 2012) withthe bones of children both smaller andless dense than those of adults, there-fore, leaving them more vulnerable todecay, ease of dispersion and loss. Childremains are easier to disarticulate and re-move by scavenging which can result inloss of elements (Waldron 1987; Mortonand Lord 2002; 2006) from both archae-ological and forensic contexts. There isvariation in the preservation of differentbones. The bones most vulnerable to de-struction are thought to be those witha high proportion of cancellous material,such as the sternum, vertebrae, ribs andthe epiphyses. It has been thought thatthe lumbar vertebrae are the least andthe cervical the most affected by soil ero-sion (Mays 1991). This may also dependon the position of the body during buri-al, and if grave intercutting occurred. Ac-cording to Mays (1991), the small bonesof the hands and feet are almost alwayspoorly represented, whilst bones witha high proportion of cortical bone, suchas the skull, mandible and long bones ap-pear to be less affected by preservation.A similar pattern was reported by Wal-dron (1987) on a study of West TenterStreet, London, who also pointed outthat this pattern of preservation is not

necessarily the same for all sites. Howev-

er, similar ndings were reported by In-gvarsson-Sundstrm (2003) from Asine,in Greece. Von Endt and Ortner (1984)have shown that rates of decay are in-versely proportional to the bone size.They found when bones of different sizeswere kept in water at constant tempera-ture; nitrogen is released at a rate whichis inversely proportional to bone size.Any weakening of the protein-mineralbonding of bone will enhance its degra-dation. Groundwater and its dissolvedions can penetrate bone, and bone size,both the external and internal surfacearea (Porosity), available to groundwa-ter is important in bone breakdown (VonEndt and Ortner 1984).

Porosity is an important factor fordiagenetic change in bone. There is anincrease in porosity as a result of min-eral dissolution. Chaplin (1971) notedthat the rate of dissolution is dependenton the porosity of the skeletal tissue, asmore porous tissue decays more rapidlythan less porous tissue. This is impor-tant for non-adult bone as it has beenshown that non-adult remains are moresusceptible to diagenetic contamination(Von Endt and Ortner 1984; Zapata etal. 2006; Hanson and Buikstra 1987)and this can be from the surroundingsoil. Amour-Chelu and Andrews (1996)found that a chalk environment was notfavourable for bone preservation at Over-ton Down, where surface modication ofnon-adult remains occurred within a fewyears due to their porous nature. Thepore structure, which can be dened asthe distribution of porosity for a givenpore radius, can inuence the amountof diagenesis. An increase in the rate ofmineral dissolution process, will lead togreater porosity (Nielsen-Marsh 2000).Hedges and Millard (1995) have high-

lighted pore structure of being of cen-


8/12/2019 anre-2013-0007

3/27

Preservation of non-adult skeletal remains 25

tral importance when modelling bonemineral loss. Pore structure governs theinternal surface area which is availablefor solid solution reactions. It also de-termines the rate at which groundwa-ter can ow through the bone, and therate at which diffusion can take place.Pore size also determines which poreswill be lled with water and which willbe empty, and so controls which parts ofbones will interact with soil water. Butaccording to Nicholson (1996:523) whoidentied bone density as an importantvariable, but stressed that bone size wasalso of importance and that it is unclearat what point bone size becomes more impor-tant than bone density...in infuencing boneloss. Bone mineral density (BMD) re-ects the degree of mineralisation of theorganic bone matrix, and this varies inevery bone. BMD increases with growthand eventually reaches a plateau in ear-ly adulthood and subsequently decreas-es with advancing age. BMD is affectedby many factors, including age, genetics,sexual maturation, physical activity anddietary calcium (Maynard et al. 1998).A number of studies have explored bonedensity in relation to child health andgrowth in past populations (Bennikeet al. 2006; McEwan et al. 2005). Morerecently, Djuri et al. (2011) found thatthe density of the femur was greater thanthat of the bula, due to its function asa weight-bearing bone. They also attrib-uted the poor preservation of infants intheir sample to bone density. In a study ofthe proximal femur and radius from twosites in England, Manifold, (forthcom-ing), found that there was an increase inBMD in infancy (01 years) in both thefemur and radius, followed by a decreasein early childhood (28 years), followedby an subsequent increase in late child-

hood (915 years).

Pathological conditions and injuriesare known to speed up the decomposi-tion of buried bone. When bone is dam-aged through trauma or as a result ofillness, it is easier for micro-organismsto enter; also the same may be said ofthose individuals with infectious diseas-es and blood poisoning. When there isa breakdown of bone in life such as withmetabolic disease, this can have an effecton the rate of preservation (Henderson1987; Breitmeier 2005). Rickets is causedby vitamin D deciency in children, pre-venting calcium from being deposited inthe developing cartilage as well as in thenewly formed osteoid, which impedesbone mineralisation. The macroscopicappearance of rickets in non-adults tendto be long bone bending deformitiesand metaphyseal swelling. However, incases of active rickets there is increasedporosity of bone surfaces in particularthe cranium and the growth plates. Thisincreased porosity can lead to the boneappearing to dissolve in the burial en-vironment, which can make recovery ofremains difcult. Another metabolic dis-ease which is not frequently diagnosed isscurvy, a condition caused by the lack ofvitamin C in the diet. This condition alsoleads to an increase in porosity in non-adult skeletons. Conditions such as thiscause a decrease in the mineralisation ofbone and this lack of mineralisation canbe misinterpreted as poor preservationrather than disease (Lewis 2010).

Extrinsic factors should also be con-sidered alongside intrinsic bone preser-vation. The presence of groundwater isimportant, especially in relation to poros-ity of bone. Hedges and Millard (1995)dened three hydrological environments:diffusive, recharge and ow. The diffusiveregime refers to an environment where

movement is limited, in waterlogged


8/12/2019 anre-2013-0007

4/27


conditions or where soils are not perma-nently saturated. With a recharge regimebones go through wetting and drying cy-cles, and as a result, porosity increasesand the formation of large pores whichincreases the effects of the water cycle.Finally, in the ow regime the presenceof bone buried in such an environmenttends to depend on the volume of water,(i.e rainfall and seasonal factors) (Hedg-es and Millard 1995). Groundwater isthe medium for all processes such asrecrystallisation, dissolution, hydrolysis,microbiological attack and ion-exchangeto take place (Nielsen-Marsh 2000). Ingeneral, bone buried in soil where watermovement is limited and calcium andphosphorous concentrations are high,has the potential to survive for an indef-inite period. Where water movement isgreater there tends to be greater disso-lution, and therefore, less well-preservedbones, both macroscopically and micro-scopically (Nielsen-Marsh and Hedges2000). Unfavourable geological condi-tions can have an impact on what bonesare likely to survive, but how much inu-ence this has on sites within the UK andskeletal remains remain unclear. Preser-vation of bone will vary considerable, notonly from soil to soil, but also from onearea of burial to another. Environmentsaffect bone in different ways. In acidicenvironment, which can consist mostlyof podsols, these soils tend to be abun-dant in northern England and Scotland,where there is a tendency for the soilsto be thin, acidic and wet, which may ormay not have a negative impact on bonepreservation (French 2003; Henderson1987). On the other hand, many peat en-vironments have revealed excellent pres-ervation due to the acidic nature of thesites, due to the lack of microbial attack

(French 2003). In a more alkaline envi-

ronment, which consists of calcareoussoils can result in mixed preservation, ifremains are recovered from this soil typeand have a high pH, they tend to be ingood condition (Brothwell 1981; Ferllini2007), and these soils tend to be foundin East Anglia and eastern and southwestEngland. In soils of a neutral pH, therecan be varied conditions, these soils arewell-drained and mostly located on thegravel and chalk areas of southern Eng-land. An increase in biological activityleads to a breakdown of organic matter,which results in a well-mixed, aeratedsoil and can lead to poor preservation(French 2003). Locock et al., (1992)found, that soil pH was not the maincontrolling factor in the preservation ofburied bone. Some demineralisation ofbone may occur as a result of the actionof organic acids released during decom-position of the soft tissues, and thereforepresent in the soil where the bones areexposed (Child 1995). Overall, the liter-ature has produced some contradictionsas to what environment is best for bonepreservation.

Other factors such as ora and fauna,plant roots and human impact shouldalso be bore in mind. Flora and fauna canattack bone directly resulting in dam-age and destruction of bone tissue, butalso indirectly resulting in scattering andbreakage of bone (Henderson 1987). In-sects can destroy human remains, theirinuence varies with conditions of burialand factors such as season, latitude andaltitude (Erzinclioglu 1983). Snails andother mammals can prey on bones, de-stroying and or alternating them whichcan lead to suggestions of pathology(Henderson 1987). Plant roots can leadto marks which resemble pathologicalconditions and thus, cause misinterpre-

tation of disease (Wells 1967). Large


8/12/2019 anre-2013-0007

5/27


roots leave indentations on the surface ofbones and often the roots grow throughthe bones leaving holes which can bemisinterpretation as ante-mortem in-juries. Roots can creep into bones andexert strong pressure on the bone walls,eventually causing fragmentation. Theycan also cause the dissolution of min-eral components of bones by excretinghumic acids. Lyman (1996) describedroot etching which results in erosionof the cortical surface and can lead tocomplete dissolution of the bones. Thetreatment of the body after death canhave signicant impact on what skeletalelements are recovered. With regard tothe remains of children this is particu-larly important as there is the commonperception that graves belonging to theyounger individuals tend to be shallow orpit graves, which can be easily exposed toplough damage, thus resulting in the lossof remains, especially those of infants.This has been observed at a number ofcemeteries. At the Roman site of Can-nington in Somerset, the graves of theinfants had a greater tendency towardsshallow graves, whereas the graves ofthe older children were similar in depthto the adults (Rahtz et al. 2000). Scull(1997) observed at Watcheld cemeteryin Oxfordshire that infants and young

children were interred in shallow gravesand those burials recovered were withinor at the base of the ploughsoil. The pur-pose of this paper is to examine the bonepreservation using three preservationalindices: Anatomical preservation index(API), Qualitative preservation index(QBI) and the Bone representation in-dex (BRI) of a number of early medievalnon-adult skeletal remains from differentgeographical and geological locations inthe UK.

MaterialsA sample of 421 skeletons from fourearly medieval archaeological sites ofdifferent geographical locations withinthe United Kingdom were studied (Ta-ble 1; Fig. 1).The sites included that ofEdix Hill, and Great Chesterford, bothin Cambridgeshire. The Scottish site ofAuldhame, East Lothian and the Welshsite of Llandough. Each of the sites willbe discussed in turn.

Edix Hill, CambridgeshireEdix Hill is situated on the western edgeof Barrington parish close to the villageof Orwell, which lay 12km south-westof Cambridge (Malim and Hines 1998).The site was dated from the sixth to the

Table 1. Archaeological sites studied

Period SiteTotal no. of non-

adult skeletonsGeology and pH Reference

Early MedievalGreat ChesterfordCambridgeshire, UK

82 Neutral Waldron (1988)

Early MedievalEdix HillCambridgeshire, UK

41 Chalk Duhig (1998)

Early MedievalAuldhame, EastLothianScotland

72 Alkaline Melikian (2005)

Early MedievalLlandoughSouth Wales

226 WaterloggedLoe and Rob-

son-Brown (2005)

Total 421


8/12/2019 anre-2013-0007

6/27


seventh centuries. The cemetery of EdixHill was situated on chalk Knoll sur-rounded by lower lying claylands (Gaultclay) which is underlying geology of thearea that had been exposed through lo-calised erosion of the chalky upper de-posits. The burials at Edix Hill weregenerally shallow and mostly comprisedsingle interments with only a few gravescontaining more than one individual.There was little patterning in the orien-tation of the graves and it would appearthat topographical factors were of moreimportance (Malim and Hines 1998).A concentration of non-adult burials wasapparent in the area on the brow of theknoll, which may indicate a particulararea of burial for children of all ages, as

both infants and adolescents were buried

here (Malim and Hines 1998). Other-wise, the graves of the children appear tohave been evenly spread out across thecemetery. The remains were damagedby agricultural processes as a result ofthe shallowness of the graves. The to-tal number of individuals recovered was148, forty-six of which were children.

Great Chesterford, CambridgeshireThe site of Great Chesterford (AD4101065) lies on the gravel terraces ofthe east bank of the river Cam, southof Cambridge city (Evison 1994). Thetown of Great Chesterford is approx-imately 15km south east of Edix Hill.Great Chesterford was an Anglo-Saxoncemetery built upon a Romano-Britishextramural cemetery (Evison 1994). Thetotal number of individuals recoveredwas 167, eighty-three of which werenon-adults. The non-adults were mostlyburied in single graves, although therewere three multiple graves. The graves ofGreat Chesterford lie in one of two direc-tions, some with their head to the south(south-north graves) and some with thehead to the west (west-east graves). Ori-entations were recorded for fty-eight ofthe non-adult graves. Most of the non-adults were buried south-north (62%;25/40).

Auldhame, East LothianThe site of Auldhame, East Lothian wasuncovered in 2005 by AOC archaeologywhere 260 individuals were recovered,and a further sixty-six burials indentiedbut were left in situ. The multi-phasedremains of a chapel were also recov-ered. Four phases of activity indentied Phase One (AD 6509501000); PhaseTwo (c. AD 9501200); Phase Three (c.AD 12501450) and Phase Four (AD

14701680) (Melikian 2005). The re-

Fig. 1. Location of sites


8/12/2019 anre-2013-0007

7/27


mains mark the site of a previously un-known medieval cemetery, and lie withinthe possible promontory fort of Seacliff.It is not known when a settlement atAuldhame rst appeared, but discoverieswithin the locality, such as the prehistor-ic round cairn at St Baldreds Cradle andIron Age burials at Greghans Cave, sug-gest occupation from at least the BronzeAge (Hindmarch and Melikian 2006).All burials were supine and extended,with most following the west to eastalignment with the head at the westernend. An isolated group of juvenile bur-ials was discovered directly to the westof the building which had alignments ofsouth-west to north-east. This may in-dicate inter-cutting of later graves andthe avoidance of in situ burials. A total of78 non-adults were recovered (Melikian2005).

Llandough, South WalesThe site of Llandough lies in the northof Penarth on sloping ground near thecrest of an escarpment which overlooksthe estuary of the river Ely (Holbrookand Thomas 2005). The excavation ofthe burial ground was undertaken byCotswold Archaeological Trust in 1994ahead of development. The excavationsarea lay to the north of the church yardwall and extended to the edges of the es-carpment. Within this area 1026 graveswere recovered. There were 814 articulat-ed skeletons and 212 disturbed skeletonsrecovered. Of these 226 were non-adults.Many of the skeletons were buried in veryshallow graves and there was evidenceto suggest activity which post-dated thecemetery which had truncated much ofthe site. During excavation burials weredivided into three areas. Area I was situ-ated in the south of the cemetery, which

included burials that were contained

within a possible curvilinear boundarywhich was indicated by the line of bur-ials on a north-east to south-west align-ment (Loe 2003). Areas II and III lay tothe west and north of Area I. Burials inArea II lay further to the west outside thelimits of the excavation. Area III was themost extensively used part of the ceme-tery. The burials were aligned east-west.This area contained a large proportionof infant and non-adults remains, whichwas clustered into two distinct groups;one which was central and the other inan adjacent area to the north. It is likelythe burials in Area I relate to the monas-tic community which was established inthe 6thcentury. This area of the cemeterywould have included the monks and layaristocracy (Davies 1982). The Areas IIand III are thought to comprise the laypopulation who were afforded the rightto be buried in monastic cemeteries fromabout the 6thcentury, this would accountfor the distribution and the majority ofburials.

Methods

Age-at death

The age of death of the non-adults wasassessed using the dentition (Moor-rees, Fanning and Hunt 1963ab) longbone lengths (Ubelaker 1989), and bonedevelopment (Buikstra and Ubleaker1994). The foetal remains were agedusing long bone lengths (Scheuer, Mus-grave and Evans 1980) and pars basilars(Scheuer and MacLaughlin-Black 1994).Of the 421 skeletons studied, 376 couldbe aged accurately. The skeletons wereplaced into ve age groups: < 40 weeks,01.5 years; 1.64.5 years, 4.610.5 years

and 10.617.0 years (Table 2).


8/12/2019 anre-2013-0007

8/27


Bone Preservation

Anatomical preservation index (API)The Anatomical Preservation Index(AP1, Bello et al. 2006) expresses theratio between the scores of preservation;the percentage of bone preserved for eachsingle bone, and the skeletons total ana-tomical number of bones. Each individu-al bone was then categorised and classedaccording to the six classes including thebones that were absent (Table 3).

Qualitative bone index (QBI)The state of preservation of the corti-cal surfaces were evaluated using theQualitative Bone Index (QBI, Bello et al.2006), being the ratio between the soundcortical surfaces and the damaged surfac-es of each bone. The cortical surface ofeach bone was examined and a class ofpreservation, which was most appropri-ate, was applied.

Bone preservation index (BRI)The Bone Representation Index (BRI)was devised by Dodson and Wexlar,(1979) and measures the frequency ofeach bone and bone type in the sample.It is the ratio between the actual num-ber of bones removed during the excava-tion and the theoretical number of bonesthat should have been present accordingto the minimum number of individuals(MNI) in the sample. Using the skeletalinventory each bone was scored as ab-sent or present. The dentition was scoredon an absence or presence basis, takinginto account age. Both the deciduous andpermanent teeth were recorded.

Intra-observer and inter-observererror

The intra-observer error is the error be-tween two measurements taken at twotimes by the same observer on the samesample and using the same methods ofmeasurement. The inter-observer erroris the error between two observers onthe same sample using the same criteria.The scores of preservation for both theanatomical preservation index (API) andthe qualitative bone index (QBI) wereestimated on sixty-eight bone elementsof two skeletons by the author and an-other experienced osteologist to test forerrors. The scores of preservation for the

API and QBI were estimated by the au-

Table 2. Age at death

Site

Age at death

Total

8/12/2019 anre-2013-0007

9/27


thor and produced a P value equal to 1.Therefore, the difference was not consid-ered signicant. The inter-observer errorwas estimated and produced a P value of0.883, which means there was no signif-icant difference between the measure-ments taken.

Statistical analysisAll data for each of the preservationscores was entered into an excel da-tabase and analysed according to thefollowing: (i) difference between bonepreservation expressed in terms of(API) and (QBI) and site (location) bothnationally and locally, (ii) difference be-tween bone preservation and age of thenon-adult.

Trends were analysed using ChiSquare statistical test to test the null hy-pothesis that there was no difference be-tween trends (Shennan 1997). A signif-icance level of 1% (p

8/12/2019 anre-2013-0007

10/27


the regional differences were analysed toobserve if any differences were present.It was found that signicant differenceswere present between each class of pres-ervation and each site (Table 5). Therewere considerably more poorly preservedbones at the sites of Edix Hill and GreatChesterford, in all classes except 3 and 6demonstrating that differences can existonly between sites which are from thesame geographical location, in this case,only 15 km apart. The Scottish site ofAuldhame was compared to the Englishand Welsh sites to assess if there is anydifference in preservation between north

and south. By considering the percentage

of bone in each class of preservation, thesame pattern emerged with large per-centages of bone elements in class oneand subsequent decrease as the preser-vation classes increased. This was sig-nicant in the number of bones presentbetween all sites and classes. This wouldalso indicate that there is a difference inbone preservation and geology of eachsite, and that the preservation of bonecannot be classied according to a par-ticular area or region. The same patternof preservation was followed at sites inEngland and Wales. It was hypothesisedthat the soils in the north of England

and Scotland were less suitable to good

Table 4. Number and percentage of bones preserved (API) at each class and age group for Auldhame, Llan-dough, Great Chesterford and Edix Hill

Age GroupClass 1 Class 2 Class 3 Class 4 Class 5 Class 6

n (%) n (%) n (%) n (%) n (%) n (%)

Auldhame

< 40 weeks 195 (72) 15 (5) 9 (3) 16 (6) 29 (11) 8 (3)

01.5 years 996 (50) 140 (7) 68 (3) 53 (3) 74 (4) 29 (1)

1.64.5 years 318 (40) 69 (9) 41 (5) 29 (4) 54 (7) 33 (4)

4.610.5 years 936 (37) 43 (6) 119 (5) 120 (5) 208 (8) 174 (7)

10.617.0 years 222 (44) 23 (5) 25 (5) 26 (5) 38 (8) 5 (1)

Llandough

< 40 weeks 301 (60) 18 (4) 8 (2) 0 5 (1) 8 (2)

01.5 years 1573 (52) 343 (11) 66 (2) 32 (1) 21 (1) 5 (0.1)

1.64.5 years 1238 (51) 240 (10) 59 (2) 33 (1) 34 (1) 8 (0.3)

4.610.5 years 1147 (48) 270 (11) 111 (5) 50 (2) 37 (1) 13 (0.5)

10.617.0 years 567 (47) 120 (10) 36 (3) 44 (4) 38 (3) 11 (1)

Great Chesterford

< 40 weeks 1240 (79) 125 (8) 28 (2) 16 (1) 34 (2) 121 (8)

01.5 years 1257 (48) 179 (7) 55 (2) 43 (2) 92 (4) 92 (4)

1.64.5 years 373 (41) 71 (8) 26 (3) 31 (3) 46 (5) 65 (7)

4.610.5 years 280 (40) 44 (6) 17 (2) 20 (3) 35 (5) 80 (11)

10.617.0 years 73 (53) 9 (7) 9 (7) 7 (5) 14 (10) 24 (18)

Edix Hill

< 40 weeks 40 (59) 10 (15) 7 (10) 3 (4) 6 (8) 2 (3)

01.5 years 71 (35) 8 (2) 8 (2) 7 (3) 21 (10) 21 (10)

1.64.5 years 587 (53) 82 (7) 36 (3) 38 (3) 51 (5) 29 (3)

4.610.5 years 259 (37) 69 (10) 33 (5) 32 (4) 46 (6) 37 (5)

10.617.0 years 666 (42) 100 (6) 34 (2) 47 (3) 111 (6) 120 (7)


8/12/2019 anre-2013-0007

11/27


bone preservation, due to the tendencyfor soils to be wet and acidic which canlead to poor bone preservation. With thesoils of the south of the United Kingdomhaving a more desirable affect on humanbone preservation.

Age differencesIn order to assess if age has an impacton the state of preservation, each of thesites and age groups were analysed (Ta-ble 4 and Table 5). No signicant differ-ences were observed at any age groupbetween the poorly preserved remainsand those exhibiting excellent preser-vation at the sites of Llandough andGreat Chesterford. At the site of EdixHill a signicant difference was notedbetween < 40 weeks and 0-.1.5 years atclass 1 (2=7.38, p=0.05, df=1), class2 (2=13.0,p=0.001, df=1) and class 3(2=9.1, p=0.05, df=1). At Auldhamea difference was noted between the ages

of 4.610.5 years and 10.617.0 years

at class 1 (2=5.1, p=0.05, df=1) andin the younger individuals (

8/12/2019 anre-2013-0007

12/27


ences were recorded between the sites ofLlandough/ Great Chesterford (2=7.1,

p=0.05, df=1; 2=10.3, p=0.05, df=1respectively), and between the sites ofLlandough / Auldhame (2=4.0,p=0.05,df=1;2=6.8, p=0.05, df=1 respective-ly). In the older children (10.617.0years) differences in good and poorpreservation were also noted betweena number of the sites, Auldhame/EdixHill (2=4.4,p=0.05, df=1), Auldhame/Great Chesterford (2=10.8, p=0.05,df=1), Llandough/ Great Chesterford(2=11.6, p=0.001, df=1), Llandough/Edix Hill (2=4.8,p=0.05, df=1).

Qualitative Bone Index (QBI)

Site differencesThe preservation of the cortical bone sur-faces was assessed according to the sixclasses of preservation, where Class 1 de-notes that the cortical surface was com-

pleted eroded, Class 2 that up to 25% ofthe surface was preserved, Class 3 thatup to 50% of the bones surface was pre-served, Classes 4 and 5 were recordedwhen 75% of the cortical surface was in-tact and Class 6 was recorded when thebones surface was completely preserved.Overall, the majority of the sites had atleast 50% of the bone surfaces present(i.e., high percentage of Class 3 preserva-tion) (Fig. 3; Table 6). At the northern siteof Auldhame, the poorly preserved corti-cal bone at classes 13 was signicantwhen compared to Edix Hill (2=149.29,

p=0.001, df=1). Thus indicating a geolog-ical inuence. Cortical bone preservationalso differs across a small geographicalarea, for example at Edix Hill and GreatChesterford, the poor cortical bone pres-ervation at Edix Hill was due to the af-fects of chalk and was signicant at Class1 (2=96.2,p=0.001, df=1) and Class 2(2=404.7,p=0.001, df=1) (Table 7).

30

25

20

15

10

5

0Class 1 Class 2 Class 3 Class 4 Class 5 Class 6

Auldhame Llandough Edix Hill Great Chesterford

%

ofbonesineachclassofpreservation(QBI)

Fig. 3 Percentage of bones in each class of preservation (QBI)


8/12/2019 anre-2013-0007

13/27


Age DifferencesAge was also assessed in relation to thestate of preservation of the cortical sur-faces (QBI) of the remains. The age ofthe non-adults at each site were com-pared statistically to observe if any differ-ences existed between the sites. No dif-ferences were observed at the four sitesor for each age group for each of the pres-ervation classes. Neither were there anydifferences in good and poor preserva-tion noted between the sites at each agegroup. The only difference recorded wasbetween Auldhame/Edix Hill at 4.610.5

years (2=6.8,p=0.05, df=1).

Grave depth and age at GreatChesterford and Edix Hill

In order to assess whether non-adultsare poorly preserved because of shallowburial depth, burial depth, age, API andQBI were analysed for the sites of GreatChesterford and Edix Hill. A differencewas found between grave depth, corti-cal surface preservation and age, at theages of 01 years and 14 years (2=9.3;

p=0.005, df=1) and a signicant dif-ference was also found at the ages of510 years and 1117 years (2=24.8;

p=0.001, df=1). There were no signif-

icant differences between the API, age

Table 6. Number and percentage of bones preserved (QBI) at each class and age group for Auldhame, Lla-ndough, Great Chesterford and Edix Hill

Age GroupClass 1 Class 2 Class 3 Class 4 Class 5 Class 6

n (%) n (%) n (%) n (%) n (%) n (%)

Auldhame

< 40 weeks 0 47 (24) 32 (17) 15 (8) 5 (3) 0

01.5 years 1 (0.05) 122 (6) 147 (7) 86 (4) 44 (2) 1 (0.05)

1.64.5 years 2 (0.2) 87 (9) 110 (11) 63 (6) 28 (3) 3 (0.3)

4.610.5 years 1 (0.04) 153 (6) 160 (7) 162 (7) 210 (9) 41 (2)

10.617.0 years 0 53 (7) 66 (9) 37 (5) 15 (2) 5 (1)

Llandough

< 40 weeks 0 10 (5) 43 (22) 2 (1) 0 0

01.5 years 4 (0.1) 58 (2) 270 (10) 33 (1) 3 (0.1) 0

1.64.5 years 5 (2) 58 (2) 188 (9) 61 (3) 0 0

4.610.5 years 20 (1) 71 (3) 268 (12) 90 (4) 11 (1) 0

10.617.0 years 2 (0.2) 5 (0.4) 90 (8) 76 (7) 49 (4) 0

Great Chesterford

< 40 weeks 24 (3) 84 (10) 111 (13) 0 0 2 (0.2)

01.5 years 18 (1) 135 (3) 381 (24) 26 (2) 1 (0.06) 0

1.64.5 years 20 (1) 45 (3) 246 (16) 58 (4) 1 (0.06) 0

4.610.5 years 7 (0.1) 11 (1) 153 (15) 69 (7) 9 (1) 0

10.617.0 years 0 3 (1) 57 (28) 2 (1) 1 (0.06) 0

Edix Hill

< 40 weeks 0 5 (10) 19 (39) 7 (14) 0 0

01.5 years 0 34 (7) 77 (15) 22 (4) 2 (0.4) 0

1.64.5 years 12 (3) 54 (14) 73 (19) 18 (5) 3 (1) 0

4.610.5 years 33 (5) 59 (8) 91 (13) 38 (5) 20 (3) 0

10.617.0 years 53 (5) 157 (10) 156 (10) 41 (3) 1 (0.06) 0


8/12/2019 anre-2013-0007

14/27


and grave depth at in the younger indi-viduals, but a signicant difference wasfound in the older children, 510 yearsand 1117 years age group (2=52.6;

p=0.001, df=1).

Bone frequencies at each siteThe number and percentages of eachbone from both the cranium andpost-cranium was calculated for each site(Table 8). A total of sixty-eight bonesfrom each skeleton, including left andright. The bones of the skull tend to varyin their preservation. Here, the cranialbones most frequently preserved includ-ed the temporal, occipital, parietal andmandible. The temporal bone was rep-resented at Auldhame (67; 48% of skel-etons), at Great Chesterford (76; 49%),Llandough (23; 27%) and Edix Hill (48;58% respectively). This may be due toits relatively high density, especially thepetrosa portion. The occipital bone was

well-represented at Auldhame (45; 64%)

and Great Chesterford (45; 58%). Thepars basilaris was especially abundantat all sites, this is an important elementin that it can facilitate in the ageing ofinfants in the absence of dental remains.The parietal and frontal bones tend tobe present, however, they are usually re-covered in fragmented form, this may beas a result of burial position (i.e supine)and it is also vulnerable to plough andexcavation damage. The mandible andmaxilla were less well-represented, withall samples having less than half pres-ent. This is surprising as the mandible isdense and compact in nature and is oftenused in the estimation of MNI evalua-tions (Brezillion 1963). The small bonesof the face, for example the lacrimal andethmoid, tend to be poorly representedat all the sites. This is likely due to thefragmentation of the skull and also thesebones can be difcult to recognize duringexcavation. The small bones of the ear

(malleus, incus and stapes) when recov-

Table 7. Statistical analyses of bones per each class of qualitative bone preservation (QBI) between all thesites

ClassAuldhamevs Edix Hill

Auldhamevs Great Ches-

terford

Edix Hillvs Great Ches-

terford

Llandoughvs Great Ches-

terford

Llandoughvs Edix Hill

Llandoughvs Auldhame

Class 12=149.3

df=1p=0.001

2=24.6df=1

p=0.001

2=96.2df=1

p=0.001

2=22.6df=1

p=0.001

2=206.6df=1

p=0.001

Class 22=181.3

df=1p=0.001

2=404.8 df=1

p=0.001

2=217.1df=1

p=0.001

2=196.4df=1

p=0.001

Class 32=40.9

df=1p=0.001

2=6.8df=1

p=0.001

2=12.9df=1

p=0.001

2=16.1df=1

p=0.001

Class 42=26.2

df=1

p=0.001

2=302.4df=1

p=0.001

2=69.9df=1

p=0.001

2=6df=1

p=0.025

2=106.7df=1

p=0.001

Class 52=118.5

df=1p=0.001

2=494.4df=1

p=0.001

2=31.1df=1

p=0.001

2=6.9df=1

p=0.001

2=399.1df=1

p=0.001

Class 62=31.2

df=1p=0.001

2=78.9df=1

p=0.001

2=96.2df=1

p=0.001


8/12/2019 anre-2013-0007

15/27


Table 8. Frequency of bones represented at each site (n=421)

BoneAuldhame Great Chesterford Edix Hill Llandough

n (%) n (%) n (%) n (%)

Temporal 67 (48) 76 (49) 48 (58) 123 (27)

Parietal 32 (45) 75 (49) 53 (65) 114 (50)

Frontal 31 (44) 38 (49) 27 (66) 41 (18)

Sphenoid 32 (46) 30 (39) 20 (49) 43 (19)

Occipital 45 (64) 45 (58) 26 (63) 69 (30)

Zygomatic 33 (23) 31 (20) 28 (34) 22 (5)

Nasal 3 (2) 2 (1) 4 (5) 0

Vomer 4 (3) 2 (1) 2 (2) 0

Lacrimal 3 (2) 0 2 (2) 0

Ethmoid 4 (3) 3 (2) 3 (4) 0

Palatine 1 (1) 2 (1) 0 0Malleus 9 (6) 15 (10) 1 (1) 5 (1)

Incus 7 (5) 15 (10) 1 (1) 2 (1)

Stapes 1 (1) 3 (2) 2 (2) 2 (1)

Hyoid 7 (5) 0 2 (2) 0

Mandible 38 (54) 34 (44) 25 (61) 43 (19)

Maxilla 30 (43) 26 (34) 20 (49) 23 (10)

Humerus 89 (79) 107 (69) 51 (62) 140 (31)

Radius 87 (62) 74 (48) 45 (55) 78 (17)

Ulna 85 (61) 66 (43) 48 (58) 104 (23)

Clavicle 77 (55) 82 (53) 38 (46) 68 (15)

Scapula 81 (58) 70 (45) 45 (55) 86 (19)

Carpals 24 (17) 2 (1) 13 (16) 15 (3)

Metacarpals 64 (46) 30 (19) 34 (41) 38 (8)

Manual phalanges 62 (44) 30 (19) 35 (43) 43 (9)

Cervical Verts 55 (78) 42 (54) 26 (63) 70 (31)

Thoraic Verts 58 (83) 48 (62) 26 (63) 58 (26)

Lumbar Verts 50 (71) 37 (48) 24 (58) 34 (15)

Ribs 126 (90) 117 (76) 58 (71) 172 (38)

Stemum 18 (26) 10 (13) 10 (24) 10 (4)

Pelvis 113 (81) 68 (44) 58 (71) 116 (26)

Sacrum 34 (48) 15 (19) 17 (41) 36 (16)

Coccyx 1 (1) 0 0 0

Femur 102 (73) 93 (60) 62 (76) 141 (31)

Tibia 91 (65) 76 (49) 53 (65) 118 (26)

Fibula 79 (56) 61 (40) 45 (55) 102 (22)

Patella 15 (11) 1 (1) 20 (24) 10 (2)

Tarsals 36 (26) 24 (15) 33 (40) 34 (7

Metatarsals 36 (26) 20 (13) 33 (40) 43 (10)

Pedal phalanges 31 (22) 8 (5) 20 (24) 35 (8)


8/12/2019 anre-2013-0007

16/27


ered are always in excellent condition,this may be due to the protection of thetemporal bone, with the malleus and sta-pes tending to be better represented thanthe incus.

The humerus was well-representedacross three of the sites; at Auldhame(89; 79%); Great Chesterford (107;69%) and Edix Hill (51; 62%). At Llan-dough the bone was less preserved (140;31%). There was a similar pattern for theradius and ulna; as well as for the clavi-cle and scapula. The bones of the handwere poorly represented at all sites, forexample at Llandough: carpals (15; 3%),metacarpals (38; 8%) and phalanges (43;9%). The bones of the spine are normal-ly abundant in assemblages, and in thisstudy the cervical and thoracic vertebratewere the most represented (Table 8).This is in contrast to other studies (Mays2010; Bello and Andrews 2006) report-ing that the lumbar vertebrae as the mostcommonly preserved portion of the spi-nal column. The bones of the lower limbwere well preserved at all sites with thefemur being the most abundant followedby the tibia. The bones of the feet werealso poorly represented.

Bone frequencies and ageThe frequency of each of the long boneswas calculated for each age group (Ta-ble 9). The long bones are the most fre-quently recovered bone elements of thenon-adult skeletons and in this studythere was variation in numbers recoveredwhen compared to age. There were sig-nicant fewer long bones in the less than40 weeks age category, however, thismay as a result of a small sample size.The humerus was the best preserved at(44; 61%). The second age category 01.5years had a substantial number of bones

present, again the humerus was the best

represented at (125; 61%) followed bythe femur at (109; 53%) and the ulna (89;34%). There is a decrease in numbers inthose aged between one and four years,followed by an increase in those agedbetween 4.610.5 years (Table 9). In theoldest children the long bones appearedless well preserved with the bula theleast represented bone (47; 37%). Withregard to the cranial bones, the temporalbone was the best preserved, especiallyin early infancy (97; 47%), and in 4 to10 years (80; 39%). This was followedby the parietal bone (71; 34%) and (79;38%) respectively (Table 9). Contrary toa previous study by Djuric et al. (2011),the mandible was less well preserved inthis study (Table 9).

Dental frequenciesThe recovery of non-adult dental remainsis not only important for ageing but alsofor the study of stress and malnutritionin the form of dental enamel hypoplasias,but also gives clues towards diet, for ex-ample high levels of caries would indicatea diet high in carbohydrates and or sugar.The frequency of tooth type was calculat-ed for all sites (Tables 1011). The dentalelements most frequently recovered forthe maxillary deciduous teeth are molarone, canine and the lateral incisor. Thedeciduous mandibular teeth also showa similar pattern with molar one, thecanine and the central incisor most fre-quently present. In the permanent denti-tion again molar one, the canines and theincisors were most frequent for both themaxillary and mandibular teeth.

Discussion

The primary aim of this study was toocompare the bone preservation of chil-

drens skeletal remains by examining as-


8/12/2019 anre-2013-0007

17/27


semblages from four early medieval cem-eteries of different geographical locationand geology within the United Kingdom.Some scholars have argued that bonepreservation is not the dominating fac-tor in the absence of such remains fromarchaeological sites, but that culturaland religious beliefs may have a moreinuential role (Crawford 1991; Sundick1978). However, with regard to the bonepreservation, numerous contradictionsexist surrounding the best environmentfor excellent preservation. Unfavoura-ble geological conditions are often cit-ed as a cause of poor preservation, buthow much inuence this has on sitesand skeletal remains in Britain remainsunclear. The geology of the United King-dom is complex, with varying types andamounts of soils in each region (Mani-fold 2012). Therefore, preservation ofbone varies considerably, not only fromsoil to soil, but also from one place ofburial to another. Environments affect

bone in different ways. In acidic environ-

ments, which mostly consist of podsols,these soils tend to be abundant in North-ern England and Scotland, where there isa tendency for the soils to be thin, acid-ic and wet, which may or may not havea negative impact on bone preservation(French 2003). On the other hand, manypeat environments have revealed excel-lent preservation due to the acidic natureof the sites, due to the lack of microbalattack and on an accumulation of organ-ic matter, which leads to the formationof blanket bog (French 2003). In a morealkaline environment, which consistsof calcareous soils which can result inmixed preservation if remains are recov-ered from this soil type and have a highpH, then they tend to be in good condi-tion (Brothwell 1981), these soils tend tobe found in East Anglia and eastern andsouth-west England. In soils of a neutralpH, there can be varied conditions, thesesoils are well drained and mostly locatedon the gravel and chalk areas of southern

England. An increase in biological activi-

Table 9. Frequency of bones at the different age groups for the whole sample (n=376)

Bone

8/12/2019 anre-2013-0007

18/27


ty leads to a breakdown of organic matter,which results in a well-mixed , aeratedsoil and can lead to poor bone preserva-tion (French 2003). The main constitu-ents of bone, the organic part (collagen)and mineral part (hydroxyapitite), arepreserved at opposing pH levels (Mays1998). It is generally known that soilswith a neutral or alkaline pH are betterfor preservation of bone rather than acid-ic soils (Henderson 1987; Ferllini 2007),but this is not always the case. Lococket al. (1992) found that soil pH was notsaid to be the main controlling factor inthe preservation of buried bone. Somedemineralisation of bone may occur asa result of the action of organic acids

released during decomposition of the

soft tissues, and therefore present in thesoil where the bones are exposed (Child1995). Overall, it would appear that theliterature has produced some contradic-tions as to what environment is best forbone preservation. Henderson (1987:48)stated that the speed of decompositionis increased in light porous soils, whilstdense clay soils may decrease the rateof decomposition, and the deeper theburial, the poorer the preservation dueto waterlogged clay (Henderson 1987;460). However, there may be limitationsto these studies using animal bones,which may react differently to those ofthe human skeleton to soil conditions.Nicholas (1996) found acid moorland

(pH 3.54.5) was the most destructive to

Table 10. Frequency of deciduous maxillary and mandible dentition at each site (n=421)

Tooth typeAuldhame Edix Hill Great Chesterford Llandough

n (%) n (%) n (%) n (%)

MaxillaLeft Deciduous Molar 2 17 (24) 12 (29) 13 (16) 24 (11)

Left Deciduous Molar 1 18 (25) 12 (29) 17 (21) 29 (14)

Left Deciduous Canine 10 (14) 8 (19) 12 (15) 11 (5)

Left Deciduous Incisor 2 8 (11) 5 (12) 8 (10) 5 (2)


Right Deciduous Incisor 1 13 (18) 8 (19) 8 (10) 16 (7)


Right Deciduous Canine 9 (13) 7 (17) 15 (18) 15 (7)

Right Deciduous Molar 1 16 (22) 15 (36) 17 (21) 34 (16)

Right Deciduous Molar 2 14 (20) 13 (32) 18 (22) 30 (14)Mandible



Left Deciduous Canine 12 (17) 9 (22) 11 (13) 15 (7)





Right Deciduous Canine 13 (18) 7 (17) 13 (16) 15 (7)

Right Deciduous Molar 1 17 (25) 15 (36) 17 (21) 34 (16)Right Deciduous Molar 2 18 (25) 13 (32) 15 (18) 28 (13)


8/12/2019 anre-2013-0007

19/27


bone and a chalk environment (pH 7.58.9) was the most favourable. However,between these two sets of gures thereare many variables and should be usedas an indication of the extremes. Maat(1987) reported that the role of soils

in preservation may be overestimated.

This should be viewed with caution, asa study based on the decomposition ofjuvenile rates has shown that microbialactivity is a major contributor to cadaverdecomposition in soil, and it also showspersistence of cadaver in soil can be in-

uenced by the surrounding tempera-

Table 11. Frequency of permanent maxillary and mandible dentition at each site (n=421)

Tooth typeAuldhame Edix Hill Great Chesterford Llandough

n (%) n (%) n (%) n (%)Maxilla

Right Molar 3 3 (4) 6 (15) 2 (2) 1 (0.5)

Right Molar 2 10 (14) 12 (29) 4 (5) 11 (5)

Right Molar 1 14 (20) 16 (39) 11 (13) 27 (13)

Right Premolar 2 10 (14) 9 (22) 3 (4) 5 (2)

Right Premolar 1 10 (14) 10 (24) 5 (6) 10 (5)

Right Canine 10 (14) 12 (29) 3 (4) 14 (7)

Right Incisor 1 12 (17) 12 (29) 3 (4) 6 (3)

Right Incisor 2 15 (21) 17 (41) 7 (8) 21 (10)

Left Incisor 2 12 (17) 10 (24) 5 (5.6) 19 (9)

Left Incisor 1 11 (15) 10 (24) 2 (2.2) 12 (6)

Left Canine 9 (13) 11 (27) 5 (6) 9 (4)

Left Premolar 1 8 (11) 11 (27) 3 (4) 11 (5)

Left Premolar 2 10 (14) 9 (22) 1 (1) 7 (3)

Left Molar 1 16 (22) 14 (34) 10 (12) 23 (11)

Left Molar 2 11 (15) 10 (15) 6 (7) 7 (7)

Left Molar 3 2 (3) 5(3) 2 (2) 2 (1)

Mandible

Right Molar 3 2 (3) 7 (17) 1 (1) 0

Right Molar 2 11 (15) 13 (31) 8 (10) 12 (6)

Right Molar 1 14 (20) 16 (39) 12 (15) 29 (14)

Right Premolar 2 10 (14) 9 (22) 0 8 (4)

Right Premolar 1 11 (15) 13 (31) 0 11 (5)

Right Canine 11 (15) 13 (31) 5 (6) 13 (6)

Right Incisor 1 14 (20) 16 (39) 3 (4) 6 (3)

Right Incisor 2 17 (24) 17 (41) 6 (7) 24 (11)

Left Incisor 2 17 (24) 18 (44) 6 (7) 22 (10)

Left Incisor 1 14 (20) 15 (36) 6 (7) 18 (8)

Left Canine 11 (15) 13 (31) 2 (2) 13 (6)

Left Premolar 1 11 (15) 11 (27) 1 (1) 12 (6)

Left Premolar 2 10 (14) 11 (27) 2 (2) 7 (3)

Left Molar 1 16 (22) 22 (27) 7 (8) 17 (8)

Left Molar 2 9 (13) 11 (27) 7 (8) 17 (8)

Left Molar 3 1 (1) 7 (17) 1 (1) 1 (0.5)


8/12/2019 anre-2013-0007

20/27


ture and soil type (Carter et al. 2008). Ina study by Nord and colleagues (2005)on degradation of archaeological objectsand bones from prehistoric graves inSweden, it was found that the environ-ment affects preservation in three ways:rstly, the chemical environment (soilactivity) mainly affects the macroscopicappearance of bone, secondly, the micro-bial activity, composed mainly of bacteriaand fungi have a destructive affect on theorganic contents of bone and histologicalstructures. Thirdly, the inorganic mate-rial is mainly destroyed by soil activity,whereas proteins degrade at a higher pH.It would appear that calcareous soils arethe most suitable for the good preserva-tion of macroscopic structure of humanbone (Nord et al. 2005).

Geographical differences in bonepreservation

In this study all sites showed a similarpattern of preservation for both the ana-tomical preservation index (API) and thequalitative bone index (QBI) from Fig-ures 2; 3 with large percentages of bonesnot present (class1) at all sites and a lowpercentage of well preserved bones (class6) (Figures 2; 3). As for age groups, therewas also a similar pattern with largenumbers of bones not preserved (class1) for all age groups (Tables 4; 6) witha gradually decrease in percentages inclass 6. Although there was a similar pat-tern with regard to well preserved bonesat each of the sites, signicant differenc-es were noted between the sites (Tables5 and 7).

At the site of Auldhame in the north,there was evidence of shallow burialwith many of the skeletons exhibitingdamage caused by plough damage. This

was also noted at the Welsh site of Lla-

ndough coupled with the effects of thewater-logged soil which resulted in over-all poor preservation. At the site of EdixHill, many of the skeletons had their cor-tical surfaces damaged by the abrasiveaction of the chalk. This was also ob-served at the Experimental Earthworksproject at overton down in Wiltshire,southern England, where bone sampleswere buried over a long period of timeshow bone modications due to thechalk (Amour-Chelu and Andrews 1996).Chalk is pure limestone and is created bythe deposits of solid calcium carbonatein water. Soils which develop upon cal-cium carbonate are known as rendzinas,these types of soils are usually shallow,porous, well-aerated and permeable andthis can serve to protect bones in somecases (Ferllini 2007).

Evidence from many archaeologi-cal sites suggest that children are bur-ied in shallower graves than there adultcounterparts which may expose themto taphonomic processes (Ascdi andNemeskeri 1970). In order to test this,the grave depths of Edix Hill and GreatChesterford were analysed. Overall, atEdix Hill there were no major differencesfound between the ages groups, whereasat Great Chesterford a more pronounceddifference was noted. No difference wasfound between the preservation of boneand depth of burial at either Edix Hillor Great Chesterford among those agedbetween 14 years, but they did occuramong the older children. Differenceswere noted between the preservation ofthe cortical surfaces (QBI) and age atboth sites. This would suggest the depthof burial inuences the cortical surfacepreservation of childrens bones, whereasgrave depth does not appear to inuencethe amount of bone preserved (API). The

state of the cortical surface preservation


8/12/2019 anre-2013-0007

21/27


was highly signicant with regard to theclasses of preservation; this may be di-rectly related to the type of soil and pH.The cortical surface of bone elementsare directly in contact with the sedimenttype, which can have a destructive affectleading to erosion of the surfaces; thiscan limit the amount of detailed informa-tion regarding pathology of the skeleton.

Bone representation

Certain bone types tend to survive bet-ter in the burial environment, especiallywith regard to the remains of children.The cranial bones such the tempo-ral (pars petrosa), sphenoid (body andgreater wings), occipital, zygomatic,and mandible tend to be well preservedand well represented from neonate toadolescent, thus allowing the skeletonto be aged accurately. The small fragilebones of the face such as the vomer, lac-rimal and ethmoid tend to be under-rep-resented at all sites (Table 8), all bonesare present at birth, with the ethmoidossied by the seventh month of foetallife and resembles the adult morphologyat birth (Scheuer and Black 2000). Theunder-representation of such bones canhamper the study of diseases such asleprosy. Also, their absent may be due todifculties in recognizing them duringexcavation. The small bones of the ear(malleus, incus and stapes) are alwaysrecovered in excellent condition, thismay be due to the protection of the tem-poral bone, with the malleus and the sta-pes tending to be better represented thanthe incus. This may be due to the relativesmall size of such bones, which can beeasily missed during excavation. An un-der-representation of such bones hinderthe study of otitis media in archaeologi-

cal samples.

The post cranial bones which arewell-preserved and represented are thelong bones of the upper and lower limbs,ribs (especially the rst and last rib)possibly due to its anatomical positionin the body; the cervical vertebrae (es-pecially the atlas and axis) tends to bebetter preserved than that of the lum-bar. The pelvis is also well preserved andwell represented in all ages. In this studythe clavicle and scapulae were well pre-served and represented at all sites, it haspreviously reported that the scapulae ispoorly preserved due to its fragile nature(Bello et al. 2002). It was observed thatthe scapulae was often recovered in ex-cellent condition in the younger children(i.e neonate) this may be due to the sizeof the body of the scapula and its com-pact nature during the early stages ofdevelopment. When it increased in sizedue to growth, the body becomes morefragile and prone to breakage. Bones ofthe limbs (i.e femur, tibia, humerus andulna) are very well-represented in thiscurrent study. The femur is the densestbone in the body and therefore, tends tobe well preserved. In this study the ulnawas also represented in high numbers,contrary to an earlier study by Bello andAndrews (2006) which reported an un-der-representation. The frequency of theupper limb bones such as the humerusand ulna, maybe due to burial positionin the grave. However, burial positionwill differ from site to site and period toperiod. There is a differential pattern ofpreservation with regard to the smallerbones of the hands and feet. When thebones of the hands and feet are recoveredthey are usually well preserved, however,in some cases they can be misidentied.These bones are also age-dependant andmay not have ossied at the time of re-

covery or excavation.


8/12/2019 anre-2013-0007

22/27


The metatarsals and metacarpals aregenerally less well represented, as arethe phalanges of the extremities. Thecarpal bones are not present in perinatalremains, only the hamate is present at24 months and the capitate emerges ataround 35 months (Scheuer and Black2000). The metacarpals and phalangesare present before birth. The phalan-ges are often unrecognized and classedas animal bone instead and there canalso be difculty in assigning them ashuman in some circumstances (Scheuerand Black 2000; Brothwell 1981). Thetalus and calcaneus are present in per-inatal remains; these bones are the twolargest of the foot bones and are easilyrecognised. The remaining ve bonesare present from the age of one year. Thefact that they are not present in great-er numbers may also be due to loss atexcavation and washing. Another bonewhich is constantly under-representedin both adult and non-adult samples isthat of the patella. In the current studyit was the least represented bone in allsamples. This echoes the nding of Coxand Bell, (1999) who also reported theunder-representation of the patella intheir forensic case study. At birth andthe rst few years of life, the patella isentirely cartilaginous with it not takingthe adult shape until early adolescences(Scheuer and Black 2000). This couldexplain why it is difcult to recogniseand is less likely to be present in young-er individuals. The sacrum, coccyx andsternum are also under-represented;this again may be due to the fact thatthey are age dependent bones, however,this was taken in account during analy-sis. The rst segment of the sacrum isnormally the most represented part ofthe sacrum. The coccyx was absent from

all samples.

The unequal representation of certainbones can be linked to bone mineral den-sity. It has been suggested that in caseswhere more dense bones are absent couldbe due to some form of burial treatment,where bones are selected or removed forburial (Bello and Andrews 2006). How-ever, added to this the storage and cura-tion of human skeletal assemblages, asis often the case that certain elementsget lost over time, which is due to hu-man error and not preservation (Mani-fold 2010). This is particularly the casewith older collections, where curationwas minimum and non-adults were notalways deemed to warrant investigation.Depending on the time period of the col-lection, the under-representation of cer-tain bone elements maybe the result offunerary practices. This is often seen inprehistoric sites, where secondary buri-al practices took place thus causing ele-ments to be displaced. The sites studiedhere consisted of early medieval cemeter-ies where burials were complete withoutany rituals, thus leading to suggestionsthat any bone loss is due to taphonomicprocesses and or excavation techniques.The frequencies of skeletal elementsrecovered in the samples studied havea similar pattern to sites in France (Bel-lo et al. 2006) and London (Bello et al.2006).

The importance of good recoveryof dental remains of children not onlycontributes to the estimation of age inboth archaeological and forensic cases,in a survey of forensic science journals,it was found that over half of all paperspublished on the skeletal remains of chil-dren focused on ageing methods usingboth deciduous and permanent teeth(Manifold forthcoming), but also allowsthe assessment of health and disease.

One of the most commonly encoun-


8/12/2019 anre-2013-0007

23/27


tered is that of dental enamel hypoplas-ias, which can be observed as lines, pitsor grooves on the enamel surface of theincisors and canines (Roberts and Man-chester 2007) the presence of these de-fects may be a indicator of stress duringgrowth and development during child-hood. Caries are commonly encounteredin archaeological remains, and especiallyin the lower molar teeth and if presentin children may gives clues as to the typeof diet but also their presence may havelead to more serious conditions such asabscesses. Diseases such as congenitalsyphilis, which is present at birth as a re-sult of developing in the foetus secondaryto venereal syphilis in the mother (Lew-is 2007) Hutchinsons incisors, moonsmolars and mulberry molars, are dentaldefects which occur in the early stages ofthe disease (Lewis 2007). In this currentstudy the dental elements which are be-ing well-represented include the upperand lower rst molars. The molars arelarger in size with the maxillary molarshaving three roots thus allowing betterchance of recovery and preservation. Alsothe canine and the lateral and central in-cisors were well-represented across allsites. These teeth are readily identiedwhen recovered either whole or in frag-mentary form. The position of the headduring burial may be a factor in what isrecovered. If the head is placed either tothe left or right side, the dentition espe-cially from the maxilla would be betterprotected from loss. In cases where themandible is well preserved allows betterprotection of those teeth at the back (i.emolars, premolars). But also during de-velopment when the dentition has yet toerupt the small and fragile dental cuspscan be recovered.

There is a pattern to what bones are

preserved and this is indeed similar to

previous ndings but there are small dif-ferences with regard to some bones.

ConclusionsThe remains of non-adult skeletons canbe affected by the many factors involvedin bone diagensis. One of the most fre-quently cited factors as a possible causeof poor preservation is shallow depth ofgraves of children. It has been believedin the past that was the case, due to theirexposure to more taphonomic processes(Acsadi and Nemeskri, 1970), however,this is not the case here, and it was ob-served that the younger individuals wereless likely to have been affected by a rangeof taphonomic processes. It was observedat the sites of Great Chesterford and EdixHill, both of which consisted of shallowburials throughout the cemetery that theremains of the non-adults did not exhib-it a high percentage of the above tapho-nomic factors, which would be associat-ed with shallow burials. Among all sitesthere is a degree of similarity between thepreservation and representation of bones,this suggests that non-adult bones havea common pattern of preservation re-gardless of the age of the non-adult andlocation and type of site. The belief thatcertain bones will be well-preserved (i.eskull, femur, humerus) and others not so(i.e phalanges of the hands and feet), istrue in some cases, but is by no meansuniversal. In a previous study on Frenchcemeteries, Bello and colleagues (2006)suggested that human remains weremore damaged under stronger taphonom-ic pressures, therefore leading to the con-clusion that bones which have a low bonemineral content and a high percentage ofcancellous bone are more affected thanothers (Bello et al. 2006). Bone preser-

vation is a complex issue and should be


8/12/2019 anre-2013-0007

24/27


treated with caution when trying to estab-lish if and to what degree it has on theskeletal remains of non-adults. It is im-portant factor in both archaeological andforensic contexts, where many post-depo-sitional factors are active. From this studyit remains unclear as to the extent the roleof geology has on the non-adult skeleton,but the results of this study clearly showthat age is not a dominating factor in bonepreservation as previously thought.

Acknowledgements

I thank Dr. Anne Crone, AOC Archaeolo-gy, Edinburgh for access to the Auldhamematerial and unpublished data. SarahPoppy, Cambridge Archaeology Unit foraccess to Edix Hill, Dr Sonia Zakrzewski,Department of Archaeology, Universityof Southampton for access to the GreatChesterford collection. Elizabeth Walker,National Museum of Wales for access tothe Llandough skeletal collection. I thankDr Louise Loe, Oxford Archaeology foran unpublished inventory on the Llan-dough site. I also thank the editor andreviewers for their comments and sug-gestions on the original manuscript. ThisResearch was funded by the University ofReading.

Conict of interests

The author declares that there is no con-ict of interests.

Corresponding author

Bernadette M Manifold, Department ofArchaeology, School of Human and En-vironmental Science, University of Read-ing, Whiteknights, Reading, RG6 6UAe-mail address: bmmanifold@hotmail.

co.uk

References

Acsdi G, Nemeskri J. 1970. History of Hu-man Lifespan and Mortality. Budapest:

Akademia Kiado.Amour-Chelu M, Andrews P. 1996. Surfacemodication of bone. In: M Bell, PJ Flow-er, and SW Hillson, editors. The Experi-mental Earthworks Projects 19601992.Council for British archaeology researchreport 100. York: Council for British Ar-chaeology. 17885.

Bello SM, Thomann A, Signoli M, Dutour O,Andrews P. 2006. Age and sex bias in thereconstruction of past populations struc-

tures. Am J Phys Anthrop 128:2438.Bello S, Andrews P. 2006. The intrinsic pat-tern of preservation of human skeletonsand its inuence on the interpretationof funerary behaviours. In: Knsel C andGowland R. editors. The Social Archaeol-ogy of Funerary Remains.Oxford: Oxbowbooks. 113.

Bello S, Sigmoli M, Rabino-Massa E. 2002.Les processus de conservation dif-frentielle du sequelette des individusimmature. Implications sur les reconsti-tutions palodmographiques. Bull MemSoc Anthrop 14(34):24562.

Bennike P, Lewis M, Schutkowski H, ValentinF. 2005. Comparisons of child mortality intwo contrasting medieval cemeteries fromDenmark. Am J Phys Anthropol127:73446.

Breitmeier D, Graefee-Kircl U, Albrecht K,Weber M, Trger HD, Kleemann WJ.2005. Evaluation of the correlation be-tween time corpses spent in in-groundgraves and ndings at exhumation. Foren-sic Sci Int 154:21823.

Brezillion M. 1963. LHypoge des mour-nouards. Dmographie. Gallia Prhistori-ca1:5063.

Brothwell D. 1981. Digging up Bones.Ithaca:Cornell University Press.

Buckberry J. 2002. Missing presumed bur-ied? bone diagenesis and the under-rep-resentation of Anglo-Saxon children.Available at: Http://www.shef.ac.uk/~-


8/12/2019 anre-2013-0007

25/27


cassem/5/buckberr.htlm. 2002, Accessedon 10th October 2006.

Buikstra JE, Ubelaker D. 1994. Standards forData Collection from Human Skeletal Re-

mains. Fayetteville: Arkansas Archaeolog-ical Survey.

Carter DO, Yellowless D, Tibbett M. 2008.Temperature affects microbial decomposi-tion of cadavers (Rattus rattus) in contrast-ing soils. Appl Soil Ecol40:12937.

Chaplin RE. 1971. The Study of AnimalBones from Archaeological Sites. London:Seminar Press.

Child AM. 1995. Microbial taphonomy of ar-chaeological bone. Stud Conserv 40:19

30.Cox M, Bell L. 1999. Recovery of human skel-etal elements from a recent UK murderinquiry: preservational signature. J Foren-sic Sci 44:4014.

Crawford S. 1991. When do Anglo-Saxonchildren count? International Journal of

Theoretical Archaeology 2:1724.Davies W. 1982. Wales in the Early Middle

Ages. Studies in the Early History of Brit-ain. Leicester: Leicester University Press.

Djuri M, Djuki K, Milovanovi P. et al. 2011.Representing children in excavated cem-eteries: the intrinsic preservation factors.Antiq 85:25062.

Dodson P, Wexler D. 1979. Taphonomic in-vestigation of owl pellets. Paleobiology5:27884.

Erzinclioglu YZ. 1983. The application of en-tomology to forensic medicine. Med SciLaw23:5763.

Evison V. 1994. An Anglo-Saxon Cemeteryat Great Chesterford, Essex. Councilfor British Archaeology report 91. York:Council for British Archaeology.

Ferllini R. 2007. Bone scatter on chalk: theimportance of osteological knowledgeand environmental assessment. In: Brick-ley MB, Ferllini R, editors. Forensic An-thropology: Case Studies from Europe.Springeld: Charles C Thomas PublishingLtd. 21631.

Guy H, Masset C, Baud CA. 1997. Infanttaphonomy. Int J Osteoarchaeol 7:22129.

Hanson DB, Buikstra JE. 1987. Histomorpho-logical alternation in buried human bonefrom the lower Illinois Valley: implica-tions for palaeodietary research. J Archae-

ol Sci 14:54963.Hedges REM, Millard AR. 1995. Bones and

groundwater: towards the modellingof diagenetic processes. J Archaeol Sci22:15564.

Henderson J. 1987. Factors determining thestate of preservation of human remains.In: A Boddington, AN Garland and RCJanaway, editors. Death, Decay and Re-construction: Approaches to Archaeologyand Forensic Science. Manchester: Man-

chester University Press. 4354.Hindmarch E, Melikian M. 2006. BaldredsAuldhame: a medieval chapel and ceme-tery in East Lothian. The Archaeologist60:379.

Holbrook N, Thomas A. 2005. An early me-dieval monastic cemetery at Llandough,Glamorgan: excavations in 1994. MedArch XLIX: 192.

Ingvarson-Sundstrm A. 2003. Children Lostand Found: a bioarchaeological Study of

Middle Helladic Children in Asine withcomparison to Lerna. PhD Thesis, Uppsa-la University, Uppsala, Sweden.

Lewis ME. 2010. Life and death in a civitascapital: metabolic disease and trauma inthe children from late Roman Dorchester,Dorset. Am J Phys Anthropol142(3):40516.

Lewis ME. 2007. The Bioarchaeology of Chil-dren. Perspectives from Biological andForensic Anthropology. Cambridge: Cam-bridge University Press.

Locock M, Currie CK, Gray S. 1992. Chemicalchanges in buried animal bone: data froma postmedieval assemblages. Int J Osteo-archaeol2(4):297304.

Loe LK. 2003. Health and Socio-Econom-ic Status in Early Medieval Wales: AnAnalysis of Health Indicators and theirSocio-Economic Implications in an Ear-ly Medieval Human Skeletal Populationfrom the Cemetery Site at Llandough,


8/12/2019 anre-2013-0007

26/27


Glamorgan. PhD Thesis, University ofBristol, Bristol.

Lyman RL. 1996. Vertebrae Taphonomy.Cambridge University Press: Cambridge.

Nawrocki SP. 1999. Taphonomic processes inhistoric cemeteries. In: Grauer, AL. EditorBodies of Evidence: Reconstruction His-tory through Skeletal Analysis. New York,John Wiley. 4966.

Nicholson RA. 1996. Bone degradation, buri-al medium and species representation: de-bucking the myths, an experiment basedapproach. J Archaeol Sci 23:51333.

Nielsen-Marsh C. 2000. The chemical degra-dation of bones. In: Cox M, Mays S, Ed-

itors Human Osteology in Archaeologyand Forensic Science. London: GreenwichMedia. 43951.

Nord AG, Tronner K, Mattsson E, Borg GCh,Ulln I. 2005. Environmental threats toburied archaeological remains. AMBIO34(3):25662.

Maat AK. 1987. Knowledge acquired frompost-war exhumations. In: A Bodding-ton, AN Garland and RC Janaway, editors.Death, Decay and Reconstruction: Ap-

proaches to Archaeology and Forensic Sci-ence.Manchester: Manchester UniversityPress. 6580.

Manifold BM. 2012. Intrinsic and extrinsicfactors involved in the preservation ofnon-adult skeletal remains in archaeologyand forensic science. Bull Int Assoc Paleo-dont6(2):5169.

Manifold BM. 2010. The representation ofnon-adult skeletal elements recoveredfrom British archaeological sites. Child-hood in the Past 3:4362.

Manifold BM. Forthcoming. Current trendsin the study of non-adult skeletal remainsin bioarchaeology and forensic science: ananalysis of publications 20002012.

Manifold BM. Forthcoming. Estimating bonemineral density in non-adult skeletal re-mains using photodensitometry.

Maynard LM, Guo SS, Chumlea WC, RocheAF, Wisemandle WA, Zeller CM, et al.1998. Total body and regional bone miner-al content and areal bone mineral density

in children aged 818y: fels longitudinalstudy. Am J Clin Nutr 68:111117.

Mays S. 2010. The Archaeology of HumanBones. London: Routledge.

Mays S. 1991. Papers from bone taphonomyworkshop at York, September 1991. Cir-caea 9(2):5458.

Malim T, Hines J. 1998. The Anglo-SaxonCemetery at Edix Hill (Barrington A),Cambridgeshire. Council for British Ar-chaeology report 112. York: Council forBritish Archaeology.

McEwan JM, Mays S, Blake GM. 2005. Therelationship of bone mineral density andgrowth parameters to stress indicators in

a medieval juvenile population. Int J Oste-oarchaeol15:15563.Moorrees CFA, Fanning EA, Hunt EE. 1963a.

Formation and resorption of three decid-uous teeth in children. Am J Phys An-thropol 21:20513.

Moorrees CFA, Fanning EA, Hunt EE. 1963b.Age variation of formation stages for tenpermanent teeth. J Dent Res 42:1490502.

Morton RJ, Lord W. 2002. Detection andrecovery of abducted and murdered chil-

dren: behavioural and taphonomic inu-ences. In: W Haglund and M Sorg, editors.Advances in Forensic Taphonomy: Meth-ods, Theory and Archaeological Perspec-tives. New York: CRC Press. 15171.

Morton RJ, Lord WD. 2006. Taphonomy ofchild sized remains: a study of scatteringand scavenging in Virginia, USA.J Foren-sic Sci 52(3):47579.

Rahtz P, Hirst S, Wright SM. 2000. Canning-ton cemetery. Britannia Monographs Se-ries, No 17.

Roberts CA, Manchester K. 2007. The Ar-chaeology of Disease. New York: CornellUniversity Press.

Scheuer L, Musgrave JH, Evans SP. 1980. Theestimation of late fetal and perinatal agefrom limb length by linear and longarith-mic regression. Ann Hum Biol 7:25765.

Scheuer L, Maclaughlin-Black S. 1994. Ageestimation from the pars basilars of thefetal and juvenile occipital bone. Int J Os-teoarchaeol3:37780.


8/12/2019 anre-2013-0007

27/27


Scheuer L, Black S. 2000. DevelopmentalJuvenile Osteology. London: AcademicPress.

Scull C. 1997. Comment. In: J Hines Editor.

The Anglo-Saxons from the MigrationPeriod in the Eighth Century: An Ethno-graphic Perpective. Woodbridge: BoydellPress. 164.

Shennan S. 1997. Quantifying Archaeology.Edinburgh: Edinburgh University Press.

Sundick RI. 1978. Human skeletal growthand age determination. Homo 29:2849.

Ubelaker D. 1989. Human Skeletal Remains:Excavation, Analysis, Interpretation.Washington DC:Taraxacum,

Von Endt DW, Ortner DJ. 1984. Experimenteffects of bone size and temperature on

bone diagenesis. J Archaeol Sci 11:24753.

Waldron T. 1987. The relative survival of thehuman skeleton: implications for paleop-

athology. In: Boddington, A, Garland AN,and Janaway RC, editors. Death, Decayand Reconstruction: Approaches to Ar-chaeology and Forensic Science. Manches-ter: Manchester University Press. 5564.

Wells C. 1967. Pseudopathology. In: D Broth-well and AT Sandison, editors. Diseasesin Antiquity. Springeld Illinois: Thomas.519.

Zapata J, Prez-Sirvent C, Martinez-SnchezMJ, Tovar P. 2006. Diagenesis, not biogen-

esis: two late Roman skeletal examples.Sci Total Environ369:35768.

Documents

anre-2013-0007