taphonomic processes on dental microwear

Effect of Taphonomic Processes on Dental MicrowearTANIA KING,1,2* PETER ANDREWS,1 AND BASAK BOZ31Human Origins Group, Department of Palaeontology, The Natural HistoryMuseum, London SW7 5BD, United Kingdom2Department of Anthropology, University College London, London WC1E6BT, United Kingdom3Department of Social Anthropology, University of Hacettepe, 06532Beytepe, Ankara, Turkey

KEY WORDS dental microwear; taphonomic agents; Miocenehominoids; Pasalar

ABSTRACT Taphonomic processes have the potential to affect micro-scopic wear on teeth and to modify the wear patterns so as to confound dietaryreconstructions based on dental microwear which was formed during thelifetime of an animal. This study describes a series of experiments which wereconducted to simulate various taphonomic agents and to record their effect ondental microwear. Three types of experiment were carried out in order toexplain anomalous microscopic wear that had been found on the dentition ofseveral hominoid specimens from the 15 M.a. site of Pasalar in Turkey. Theeffect of two different acidscitric and hydrochloric acidon dental microwearwas investigated. Modification to microscopic wear caused by alkali (carbonatiteash) was examined in the second set of experiments. Lastly, the effect of abrasionby three different size classes of sediment from the site of Pasalarquartzpebbles (grain size varied from 2,00011,000 m), coarse sand (grain sizeranged from 5001,000 m), and medium-sized sand (grain diameters werebetween 250 and 500 m)was investigated. Results confirm previousfindings that the taphonomic modification of dental microwear is readilyidentifiable and causes the obliteration rather than secondary alteration ofmicrowear features. The experiments show that both citric and hydrochloricacid affect dental microwear but to varying degrees, whereas alkali did not causeany modification of microscopic features. The different size classes of sediment alsohad different effects on the dental microwear. The largest size sediment (quartzpebbles) polished the enamel and removed finer microwear features. The coarsesand, however, did not have any effect on the microwear. The greatest amount ofabrasion was caused by the smallest sediment particlesthe medium-sized sand.Several hominoid dental specimens from Pasalar display similar microscopic wearto the two types of acid erosion and the abrasion caused by the medium-sizedsands.Am J PhysAnthropol 108:359373, 1999. r 1999 Wiley-Liss, Inc.

During an animals lifetime, microscopicalterations to its teeth reflect the diet itingests. The examination of these wearmarks (dental microwear) has been an impor-tant tool in the reconstruction of palaeodietfor more than 20 years (for reviews seeGordon, 1988; Teaford 1988, 1991, 1994).However, after death a number of agentshave the potential to cause damage to the

enamel that is not related to the ingestion offood. Taphonomic processes are those which

Grant sponsor: The Natural History Museum; Grant sponsor:Central Research Fund (University of London); Grant sponsor:Calouste Gulbenkian Foundation.

*Correspondence to: Tania King, Human Origins Group, De-partment of Palaeontology, The Natural History Museum, Crom-well Road, London SW7 5BD, UK. E-mail: [email protected]

Received 12 March 1998; accepted 23 November 1998.

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 108:359373 (1999)

r 1999 WILEY-LISS, INC.

affect the bones of an animal after its deaththrough to fossilisation and recovery andconservation of the remains. Sedimentaryabrasion, weathering, and exposure to anacidic environment are just a few of thetaphonomic agents that have the potentialto alter or obliterate existing microscopicfeatures, and they need to be taken intoconsideration by researchers making di-etary inferences based on dental microwearanalysis. In addition, specimens may bedamaged during and after recovery. Forexample, cleaning can often affect enamel,but these types of postmortem damage arenot under investigation here.

Several experiments have been done tosimulate taphonomic processes and to docu-ment their effects on enamel and dentalmicrowear features (e.g., Gordon, 1983, 1984;Puech et al., 1985). In an investigation of theeffects of sedimentary transport on mi-crowear, Gordon (1983 and 1984) tumbledhuman teeth in four different types of drysediment and in aqueous mixtures of thesesediments. She found that microwear fea-tures were altered and in some circum-stances completely obliterated, with theamount of alteration to the microwear beingpositively correlated with the size of thesediment particles. There was no evidence,however, that new features were added toexisting microwear patterns. Puech et al.(1985) also conducted an experiment inwhich a tooth was abraded with an air-propelled stream of sand. At first the enamelbecame pitted, and then it took on a rough,eroded appearance, often with the abrasivedamage following the orientation of theenamel prisms. Striations were formed whensand particles were rubbed across the enamelsurface of another specimen. These authorsobserved that aprismatic enamel was erodedafter a tooth was tumbled in water only andthat the superficial enamel on another speci-men was dissolved, with remaining featuresbeing blunted after tumbling in seawater.When quartz was added to the seawater, thetooth became rounded, and some pieces ofenamel were chipped away from the surface.A number of striations were formed similarto those produced by the sand abrasionexperiment. As the distance travelled by thetooth in the tumbler increased, the abrasion

process became more rapid (exponentially,as the speed of abrasion was not greatlyaccelerated until after 400 km of tumbling).The total distance that the specimen trav-elled was 4,000 km, although the time takenfor this is not given the authors.

The effect of the chemical erosion of enamelwas also investigated by Puech et al. (1985)who etched a tooth with a 30% solution ofphosphoric acid for 60 sec. Initially shallowpits appeared in the enamel which, withrepeated applications of acid, then enlargedto expose a network of enamel prisms. Even-tually the prisms could no longer be distin-guished, and no features could be seen at all.

THE SITEThe Miocene site of Pasalar is situated in

northwestern Turkey 75 km southwest ofthe town of Bursa. German geologists discov-ered the site in 1969, and excavations werecarried out in 1969 and 1970, during which86 complete hominoid teeth were recovered(Andrews and Tobien, 1977). In 1983, system-atic excavations were resumed at the siteand continue to the present day (Alpagut,1990).

The deposits at Pasalar form two distinctseriesthe lower series and the upper se-riesthat are derived from two differentsources. The lower series sediments containtwo fossiliferous units. Four stratigraphicsections have been described: lower calcare-ous silt, fossiliferous sand, upper sand, andupper calcareous silt (Andrews and Alpagut,1990). The fossiliferous sand is the mainfossil-bearing unit and consists of poorlysorted sands, gravel, silt (Andrews and Alpa-gut, 1990). This unit consists of three levels:the upper level, which contains the lowestconcentrations of fossils and consists ofmainly fine sands; the middle section, whichis comprised of gravelly, fine sand and con-tains large amounts of small mammals anda few larger ones; and the lower half, whichconsists of coarse sands and gravel andcontains the greatest proportion of largemammals and hominoids (Andrews, 1995).The lower series sediments at Pasalar arelocally derived.

Two stages of fossil accumulation anddeposition have been described for the fossil-iferous sands (Andrews, 1995). The first

360 T. KING ET AL.

phase involved the accumulation of the bonesbefore burial. The evidence for this comesfrom the pattern of surface weathering whichindicates slow accumulation of bones over aperiod of time (Andrews, 1995). The fossilswere then transported with the sediments tothe fossil site, and burial occurred rapidly,probably in a matter of days (Andrews andAlpagut, 1990) in a single depositional event.The fauna is thus restricted in temporal andgeographic space.

Fifty-two mammalian species have beenidentified at Pasalar, and the assemblage isconsidered to be a natural community wheremany of the species were contemporaneous(Andrews, 1990, 1995). The hominoid speci-mens from Pasalar include three maxillae,two mandibles, 603 complete permanentteeth, 42 complete deciduous teeth, and 11postcranial elements. These specimens areconsidered to represent at least 35 individu-als and are based on material recovered upto 1989 (Alpagut et al., 1990). The number ofspecimens has increased since then, butthese lists remain unpublished at present.Two species of hominoid are probably repre-sented at Pasalar based on both metrical(upper M1) and morphological variation inthe canines and incisors (Alpagut et al.,1990). Ninety percent of the hominoid sam-ple is represented by one speciesGriphopi-thecus alpaniand the remaining 10% hasnot yet been given a species name but hasbeen assigned to Griphopithecus sp. (Alpa-gut et al., 1990).

During initial SEM (scanning electronmicroscope) examination of Miocene homi-noid (Griphopithecus alpani) specimens fromPasalar, Turkey, for the purposes of dietaryreconstruction (King et al. 1994, 1999; King,1997), features were found on some of thespecimens which differed from the mi-crowear associated with diet and/or jawmovement. The possibility was consideredthat a taphonomic process had altered themicrowear on some of the specimens.

THE PROBLEMThe occlusal phase 1 wear facet (facet 1)

(Kay and Hiiemae, 1974) on specimen H272(Fig. 1a) from Pasalar displays a number ofstriations which run parallel to each other,and the surface is rough and abraded in

appearance. This is in contrast to the usualappearance of this facet, which is character-ised by the presence of striations which runparallel to each other on a polished surface(specimen K1367) (Fig. 1b). It was suspectedthat the abrasion seen on the specimen inFigure 1a may have been the result of thepostmortem processes (for example, physi-cal abrasion by sediment) rather than causedby diet during life.

Fig. 1. Pasalar specimens suspected of having beenmodified by taphonomic processes. a: Facet 1 of speci-men H272 has a rough surface texture indicative ofpossible abrasion by sediment. Scale bar 5 50 m. b:Normal facet 1 (specimen K1367) with smooth surfacetexture. Scale bar 5 50 m. c: Specimen K1375 mayhave been etched by acid. Scale bar 5 20 m.

361TAPHONOMIC ALTERATION OF DENTAL MICROWEAR

On another specimen from Pasalar(K1375) (Fig. 1c), a number of round pits canbe seen on the surface which are arranged ina uniform manner. These pits are part of theenamel structure (enamel prisms), and theenamel appears to have been etched by acid,thereby exposing the enamel prisms. It isnecessary to know whether the apparentacid etching had occurred during the ani-mals life (i.e., through the ingestion of acidicfruits) or as a result of postmortem exposureto an acidic environment.

MATERIALS AND METHODSSix human lower molar teeth (with the

roots still attached) from the late Neolithiccave burial site of Burmegnez in southeastMalta (for details of this site see Keith,1924) were used for three taphonomic experi-ments. These were isolated, uncataloguedteeth which are housed at The Natural His-tory Museum, London. The specimens werefirst cleaned following the methods of Gor-don (1980, 1982) in order to remove any dirtthat might have obscured microscopic dam-age. This entailed gently brushing the teethwith water and detergent using a sable paint-brush, and then applying a 50% solution ofhousehold bleach to lift any organic matter,followed by a final rinsing with water.

The occlusal surfaces of the teeth wereexamined using an ISI ABT-55 scanningelectron microscope (SEM) in WET SEMmode. This type of SEM is useful for theexamination of original specimens, as thereis no need for a conductive coating to beapplied to the sample or for it to be perma-nently mounted on a stub. The WET SEMworks in the following way. While the elec-tron gun and column are maintained at ahigh vacuum (1024 torr), the specimen cham-ber is held at a lower vacuum (10211022torr) (Taylor, 1986). High energy back-scattered electrons (BSEs) are used to imagethe specimen as the increased noise (air) inthe chamber resulting from the lowervacuum interferes with the signal. The ISIABT-55 SEM is fitted with a Robinsonsback-scattered electron detector which is awide-angled scintillator-photomultiplier.This type of detector increases the signalemitted by the specimen by amplifying thenumber of electrons collected by the detector

(Bozzola and Russell, 1992). The air mol-ecules present in the chamber help to dissi-pate any electrical charge which may buildup on the specimen, which means that speci-mens do not need to be coated with a conduc-tor. This has been a very beneficial develop-ment in terms of examining valuable fossilmaterial (for full discussion of this SEMsystem and its use in the examination offossil material see Taylor, 1986). Most previ-ous microwear research has been carried outusing secondary electron (SE) imaging forwhich specimens require the application of aconductive coating. Examination of speci-mens has revealed that BSEs produce betterimages in terms of contrast, clarity, anddepth of field (King et al., 1994; King, 1997).BSEs also produce images which are morerepresentative of what is seen under a lightmicroscope; that is, they are closer to reality(Taylor, 1986). Ungar (1994) has also notedthe advantages of BSE imaging with regardto dental microwear analysis. Depending onthe type of conventional SE SEM used,differences between BSE and SE SEM tech-niques could result in variations both in theappearance of dental microwear featuresmade during an animals life and micro-scopic alterations caused by experimentalabrasion/erosion.

Before each experiment, the specimenswere examined in the SEM to provide arecord of the unaltered microscopic wear.Specimens were oriented so that the mesialedge of the occlusal surface was aligned atthe top of the viewing screen of the SEM,and magnification, which ranged from 37 to3200, varied from specimen to specimen butwas always standardised for each individualtooth. This range of magnification was usedso a general picture of the modification oflarger areas of the enamel surface could beseen as well as details of smaller patches ofthe enamel. Working distance also variedaccording to specimen but was always consis-tent for each tooth examined. An area on thehypoconid was examined for each specimenbecause it would be readily identifiable asthe experiments progressed.

Acid erosionHydrochloric acid. One specimen wasplaced in a 2.5% solution of hydrochloric

362 T. KING ET AL.

acid (HCL) (pH 0.66) in order to simulate theeffects of digestion by a predator. The toothwas left for 30 min and was then examinedusing a light microscope to check for anygross damage that might have occurred. Noalteration was apparent, and the specimenwas then placed in the acid for another hourand checked again. It was then immersed inthe acid for a further hour, rinsed withwater, and examined in the SEM. The totaltime of immersion in the acid was 2 h and 30min.

Citric acid. A second experiment was con-ducted using acidic fruit. This was to simu-late acidic etching which may have occurredduring chewing in the mouth in addition tothat which might have taken place duringdigestion by a predator. A specimen wasplaced in concentrated citric acid (lemonjuice, pH 2.16) to simulate the effects ofacidic fruit on dental microwear. The toothwas left in the acid for an uninterruptedtime of 46 h, after which it was rinsed inwater and examined in the SEM.

It should be noted that the erosion experi-ments carried out in this study were testingthe effect of acids on dental microwear with-out controlling for the level of acidity or forfactors such as the buffering effect of saliva.

Alkali erosionAs the deposits at Pasalar are alkaline, an

experiment was carried out to investigatethe effects of alkali on enamel and dentalmicrowear. A tooth was placed for a total of238 h in an aqueous solution of carbonatiteash (with a pH of 10.54) from OlDonyoLengai, a volcano in Tanzania. The speci-men was inspected under a light microscopeafter 2, 4, 8, 16, 32, 64, and 128 h to identifyany gross damage that might have occurred.As none was apparent, the tooth was exam-ined in the SEM after 168 h and 238 h.

Sediment abrasionThe effects of sedimentary transport and

deposition on dental microwear defects wereinvestigated using a commercial tumbler.Three different types of sediment were usedto examine the effects of different sizedsands: quartz pebbles (grain diametersranged from 2,00011,000 m), coarse sand

(grain size varied from 5001,000 m), andmedium sand (grain diameters were be-tween 250 and 500 m). These sedimentgrades were obtained from screened andsorted samples of the Pasalar sediments.One tooth was placed in one of the barrelscontaining each of the three types of sedi-ment. A small amount of water was added torender the sediment fluid. Each barrel wasplaced on a rotary tumbler which spun at 35revolutions per minute. The circumferenceof the barrels was 48.5 cm, and the linearvelocity was 28.3 cm/sec, or just over 1 km/h.Specimens were initially rotated for 2 h,after which they were inspected under alight microscope for any gross modificationsthat might have taken place. As none wasapparent, the specimens were tumbled foranother 2 h. As there was still no alterationapparent, the teeth were rotated for periodsof time which increased on a log2 scale: 2, 4,8, 16, 32, 64, 128, and 256 h. These dura-tions were chosen so that at the beginning ofthe experiment short time intervals wereused in order to check for rapid alteration.This gave a total number of 512 h abrasion,by which time the specimens had travelled521 km. At the end of each abrasion session,the specimens were checked under a lightmicroscope for any damage. In addition,they were examined in the SEM at fourintervals: 16, 64, 256, and 512 h. The abra-sion experiments were intended to simulatethe movement of fossils in sediment or sedi-ment which is turned over by wave action.

RESULTSAcid erosion

Hydrochloric acid. The enamel surfaceof the tooth before it was immersed in thehydrochloric acid was smooth and polishedwith the presence of microwear featuresstriations and pits (Fig. 2a). After 2 h immer-sion in hydrochloric acid, almost all of themicrowear features were removed (Fig. 2b).The acid exposed a uniform honeycomb pat-tern of enamel prisms over the entire wearsurface. It was not possible to find the exactarea on the tooth where the original SEMexamination was carried out because of theobliteration of microwear features and thealteration of the enamel surface.


Citric acid. The enamel surface of thetooth before it was placed in concentratedcitric acid was polished and displayed mi-crowear features (Fig. 2c,d). After 46 h ofimmersion in citric acid, damage to theenamel occurred around the area of exposeddentine and enamel prisms were exposed.Only the finer microwear features were re-moved by the citric acid, and the remainingdefects became deeper with sharper mar-gins (Figs. 2e,f). The alteration of the enamelby the citric acid was not so extensive as that

caused by hydrochloric acid. There was notsuch a uniform exposure of the enamelprisms in the specimen which was exposedto citric acid as compared to that which wasimmersed in hydrochloric acid, even afterimmersion for a much longer period of time(Fig. 2b,d,f ).

Alkali erosionThe enamel surface of a tooth before it was

immersed in carbonatite ash for 238 h wassmooth with microwear features present

Fig. 2. The effect of acid on dental microwear. a: Tooth before immersion in hydrochloric acid. Scalebar 5 143 m. b: Tooth after 2 h of immersion in hydrochloric acid. Scale bar 5 50 m. c,d: tooth before (c)and after (d) exposure to citric acid for 48 h. Scale bars 5 345 m. e,f: specimen before (e) and after (f)exposure to citric acid for 48 h. Scale bars 5 50 m.

364 T. KING ET AL.

(Fig. 3a). No modification to the enamel ormicrowear occurred after the specimen wasexposed to carbonatite ash (Fig. 3b). In fact,features can be seen more clearly than be-fore immersion in the alkali.

Sediment abrasionQuartz pebbles. Microwear features werepresent on the occlusal surface of the toothbefore it was tumbled with quartz pebbles(grain size 2,00011,000 m) (Fig. 4a). After64 h, the enamel had been lightly polished,and some of the finer and shallower mi-crowear features had been eroded (Fig. 4b).Further polishing occurred in the remainingtumbling sessions and is documented bySEM micrographs taken after 256 (Fig. 4c)and 512 h (Fig. 4d). The polishing anderosion that occurred in this experimentwere not extreme. Only the finer featuresthat had little depth to them were removed,giving the enamel a smoother appearance onsome areas of the enamel surface.

Coarse sand. No modification to theenamel occurred on the specimen that wastumbled in coarse sand (grain size 5001,000 m). Before the experiment, the toothexhibited microwear features over the entireenamel surface (Fig. 5a). At no stage duringthe experiment, even after 512 h of abrasionby the coarse sand, was there any oblitera-tion of the microwear (Fig. 5be).

Medium sand. The most noticeable alter-ation to the teeth used for the sedimentabrasion experiments occurred when thespecimen was tumbled with medium sand(grain size 250 with 500 m). Microwearfeatures were present on the surface of thetooth before the experiment began, withdentine exposed on the cusp tip (Fig. 6a).After 16 h of abrasion, a large area ofdamaged enamel resembling a pit appearedbelow the exposed dentine (Fig. 6b). Noother modification occurred. The damagedarea was slightly larger after 64 h (Fig. 6c),and after 256 h (Fig. 6d) it had furtherenlarged, as had the damage just below thedentine exposure. The most dramatic changeoccurred after 512 h (Fig. 6e), with extensivepitting all over the buccal face of the cusp(i.e., the left halffrom top to bottomofthe micrograph). The pits were small, withthe majority being smaller than 5 m. Thetexture of the enamel was rough and lackedany polishing at all. This contrasts with theeffect of mastication, which usually pro-duces some polishing of the enamel. Mostmicrowear features were obliterated, andthe few remaining in this area were eroded(Fig. 6f). The damaged area was approxi-mately five times bigger than it was after256 h of abrasion (see Fig. 6d) and hadcoalesced with the exposed dentine (Fig. 6g).

Heavy pitting caused by abrasion is evi-dent on the buccal portion of the cusp, and itcan also be seen inside the damaged area.However, the lingual face of the cusp, to-wards the centre of the occlusal surface ofthe tooth (i.e., to the right of the micro-graph), remained unaltered, and microwearfeatures can still be seen here. The samepattern of heavy pitting caused by abrasionon the outer areas of the occlusal surface butnot towards the centre of the tooth was seenalso on each of the cusps (protoconid, hypoco-

Fig. 3. The effect of alkali on dental microwear.Tooth before (a) and after (b) immersion in an aqueoussolution of carbonatite ash. Orientation is slightly differ-ent in the two micrographs. Scale bars 5 50 m.


nulid, metaconid and entoconid). One pos-sible explanation for this is that the longaxis of the tooth affected the way the toothrotated in the sediment and so controlled thedistribution of abrasion. The roots of thetooth were intact on the specimen used inthis experiment, and the long axis of thespecimen ran from the top of the crown tothe base of the roots, so it is possible that thetooth rotated around the long axis, withmaximum damage occurring around the out-side of the tooth.

In order to try to clarify this pattern ofabrasion, we conducted a further experi-ment. This time an extracted human rightlower molar was used, and the roots of thespecimen were removed to change the longaxis of the tooth to mesiodistal. The speci-men was tumbled for 16 h in the samequantity of medium sand and water as forthe previous experiment. This amount ofabrasion time was chosen as it was after thislength of time that the first SEM examina-tion of the specimen in the first experimenttook place. This allowed direct comparisonof the two sets of micrographs.

Before the experiment began, microwearfeatures could be seen on the enamel surfaceof the protoconid cusp (Fig. 7a,c). After 16 habrasion in medium sand, extensive abra-sive pitting could be seen (Fig. 7b,d; Table 1)with the obliteration of the microwear fea-tures. The abrasion followed the same pat-tern as the first experiment after 512 h; thatis, there was heavy pitting on the buccalfacets of the occlusal surface of the cusp butnot on the lingual face.

The nonocclusal surfaces of the extractedtooth were also investigated. The mesialcontact facet (Fig. 7e) both superiorly and atits centre and the buccal non-occusal surfaceof the protoconid (Fig. 7f) did not display thesame heavy abrasive pitting which was pre-sent on the occlusal surface.

These experiments indicate that sedimentcan cause abrasion of enamel. Table 1 sum-marises the effects of the three types ofsediment on the enamel and dental mi-crowear features after SEM examination atvarious intervals. The largest size particles(quartz pebbles) polished the enamel andremoved some of the finer and more shallow

Fig. 4. The effect of abrasion by quartz pebbles on dental microwear. Specimen before (a) and after 64(b), 256 h (c) and 512 h (d) of abrasion. Scale bars 5 50 m.

366 T. KING ET AL.

features. No alteration to the enamel ormicrowear features occurred when a speci-men was abraded with coarse sand. Thesmallest particles used in this studymedium sandcaused the most damage tothe microwear features and enamel. Theabrasion resulted in complete removal ofmicrowear features, and extensive pittingwas produced.

The differences in abrasion caused by thethree sizes of sediment may be related todifferences in the mineral composition ofquartz pebbles and sand. This might explainwhy quartz pebbles polished and removed

only some of the finer microwear features,while the medium sand caused much moreextensive removal of microwear. However,differences in the mineral composition of thesediments would not explain why only themedium sand, as opposed to the coarse sand,caused an alteration to microwear features.

Finally, a comment must be made aboutmagnification. Most microwear studies haveused magnifications for imaging specimensof between 3200 and 3500. In the presentstudy, we used a maximum magnification of3200, and this may have implications forfiner microwear features which can be seen

Fig. 5. The effect of abrasion by coarse sand on dental microwear. Tooth before (a) and after 16 h (b,64 h (c), 256 h (d), and 512 h (e) of abrasion. Scale bars 5 50 m.


Fig. 6. The effect of abrasion by medium sand on dental microwear. Tooth before (a) and after 16 h (b),64 h (c), 256 h (d) and 512 h (e). Scale bars 5 172 m. f: Detail of obliteration of the microwear features.3200. Scale bar 5 50 m. g: Coalescence of large abrasion pit with dentine exposure. Scale bar represents556 m.

368 T. KING ET AL.

only at higher magnifications. Since boththe erosion and tumbling and experimentsresulted in the removal of both fine andlarger microwear features, it can be ex-pected that fine microwear features seen athigh magnifications would also be have beenremoved. This may have occurred earlier onduring the experiments. However, it wouldbe expected that removal of finer microwearfeatures would be readily detected at highermagnifications since characteristic patternswere produced by the simulated taphonomicprocesses. In the case of the experiments

using coarse sand where no polishing oralteration was seen at lower magnifications,it might be that removal of finer featureswould have been detected at higher magnifi-cations.

Comparison with PasalarTable 2 lists all specimens from Pasalar

examined to date which appear to have beenaltered by the taphonomic processes exam-ined in this paper. One specimen from thePasalar sample (BP1312, housed at TheNatural History Museum, London) dis-

Fig. 7. The effect of medium sand on dental microwear (second experiment). a,b: Tooth before andafter abrasion. Scale bars 5 588 m. c,d: Tooth before and after abrasion. Scale bars 5 200 m. e,f: Mesialcontact facet and buccal (nonocclusal) surface of the protoconid. Scale bars 5 50 m.


played a similar type of etching pattern tothat caused by the hydrochloric acid. Thetop layer of enamel on the paracone cusp tiphas been removed, and the network of inter-connected enamel prisms has been exposed(Fig. 8a,b). The hypocone of the same speci-men also displays this pattern of prismexposure (Fig. 8c,d).

Two molars from PasalarBP36 (housedat The Natural History Museum, London)and K1375 (housed at the University ofAnkara)display a similar pattern ofenamel etching as the specimens which wereimmersed in citric acid (Fig. 8e,f). The finermicrowear features have been removed, andthere are patches of exposed enamel prisms.The features which have not been obliter-ated (especially specimen BP36) (Fig. 8f )are eroded and do not have accentuatedmargins as can be seen on the tooth whichwas etched with citric acid.

A similar pattern of pitting to that of thespecimens which were experimentallyabraded with medium-sized sand is seen intwo specimens from Pasalar. Very few mi-crowear features can be seen on specimen

C99 (housed at the University of Ankara)(Fig. 8g), and those which are present areextremely eroded. There is a general rough,pitted appearance to this specimen that issimilar to modifications caused by abrasion,although the experimentally produced alter-ation is much heavier than that seen in thefossil tooth. It is therefore likely that toothC99 from Pasalar has been abraded by sedi-ment. A second specimen from Pasalar(H272) (Fig. 1a) also has a similar aspect tothe specimens that have been abraded bymedium-sized sand but to a much lesserextent than C99. Microwear features havenot been eroded away, but the texture of theenamel appears rough and pitted and isreminiscent of the specimens that were ex-perimentally abraded by sediment.

DISCUSSIONThe purpose of the experiments reported

here was to investigate the effect that vari-ous taphonomic processes can have on micro-scopic dental wear made during an animalslifetime. Microscopic features that were sus-pected of being caused by taphonomic pro-cesses have been found on some of the fossilhominoid teeth from Pasalar, Turkey. Asthese marks could potentially confound anydietary reconstruction, it was important toascertain whether they had been producedduring the lifetime of the hominoids (i.e., bydiet) or whether they were caused by pro-cesses which took place after death.

Effect of acids on dental microwearThe experiments conducted demonstrate

that acids do affect enamel and dental mi-

TABLE 1. Effects of three sediment types on dental microwear

Abrasiontime Quartz pebbles Coarse sand

Medium sand(first experiment)

Medium sand(second experiment)

16 h Enamel polishing and fea-ture removal

No alteration Appearance of pit belowdentine exposure

Extensive pitting overenamel surface but noton lingual face of cusp


No alteration Enlargement of pit


No alteration Further enlargement of pit


No alteration Further enlargement of pitand coalescence withdentine exposure; exten-sive pitting over enamelsurface but not on lingualface of cusp

TABLE 2. List of modified specimens from Pasalar

Specimen

Erosion agents

Acid Alkali Sediment

BP36 x xBP61 x xBP65 x xBP1312 x xF184 x xK1375 x xC99 x x H272 x x

370 T. KING ET AL.

Fig. 8. Pasalar hominoid molars which have beentaphonomically modified. a,b: Acid etching of paraconecusp tip area of specimen BP1312. a: Scale bar 5 256m. b: Scale bar 5 55.6 m. c,d: Acid etching ofhypocone cusp tip region of specimen BP1312. c: Scale

bar 5 263 m. d: Scale bar 5 55.6 m. e: Acid erosionover entire occlusal surface of specimen K1375. Scalebar 5 50 m. f: Acid etching of occlusal surface ofspecimen BP36. Scale bar 5 55.6 m. g: Sedimentabrasion of specimen C99. Scale bar 5 50 m.


crowear by removal, in varying degrees, ofthe features and exposure of enamel prisms.Specimens which have been acid-etched canbe readily identified and when encounteredshould be excluded from analyses from whichinferences about past diets are made.

Hydrochloric acid caused heavy erosion ofthe microwear and enamel. Lighter etchingresulted from immersing a specimen in cit-ric acid. Both these types of erosion havebeen found in the Pasalar sample.

These two kinds of alteration can be easilyidentified in fossil samples. In the extreme,as with hydrochloric acid, almost all of themicrowear features are removed, and theenamel is etched such that the underlyingenamel prism network is exposed in a uni-form manner over the whole surface of thetooth. Puech et al. (1985) also found thispattern of prism exposure when they etcheda tooth with a 30% solution of phosphoricacid. With the lighter erosion, as seen in thecitric acid experiment, the finer microwearfeatures are removed, the margins of theremaining ones are sharpened, and thereare localised patches of prism exposure. Thispattern of removal of the more delicatemicrowear features has also been docu-mented by Teaford (1994) in an experimentthat investigated the effect of a 23 secetching of enamel with a 30% solution ofphosphoric acid.

Effect of alkali on dental microwearNo apparent alteration to the microwear

features was evident from exposure to al-kali. After the experiment had been carriedout, however, the microwear features couldbe seen more clearly than prior to the experi-ment. It may be that the alkali had acleaning effect on the tooth, resulting ingreater clarity of microwear features. Slightdifferences in the orientation may also ac-count for the variation in the appearance ofmicrowear features before and after expo-sure to alkali, since it has been found thatvariation in specimen orientation can affectthe definition of microwear features (Gor-don, 1988).

Effect of Sediment on Dental MicrowearApart from the occurrence of polishing by

the quartz pebbles, only the medium-sized

sand was found to modify the microwearfeatures within the times and degrees ofabrasion used in these experiments. Theenamel surface was roughened by extensiveabrasion pitting, and the microwear fea-tures were completely removed. This type ofabrasion has been seen in three specimensfrom the Pasalar.

It is not clear why abrasion should occurafter different periods of time in the twoexperiments using medium-sized sand. Thearchaeological tooth was modified after 16 h,as seen by the appearance of a large pit closeto the dentine exposure. However, extensiveabrasion was not seen until 512 h of abra-sion had taken place. Similar extensive abra-sion of the extracted molar occurred afterjust 16 h tumbling in medium-sized sand.Why there should be such a difference in thetime taken to cause the same amount ofdamage to these two specimens is not evi-dent. It could be related to different enamelhardness in the two specimens or the factthat one is an archaeological specimen andthe other relatively recently extracted. Inany case, although further work is needed toclarify this issue, it does not alter the factthat this size of sand particles did abradethe enamel surface and that this type ofabrasion can be recognised.

The experiments described above indicatethat although sediment has the potential toalter dental microwear patterns, the modifi-cation is in the form of obliteration of fea-tures rather than secondary alteration ofthe existing microwear patterns or the for-mation of new features. Thus, enamel thathas been abraded by sediment can be de-tected and the specimen excluded from fur-ther analysis. This conclusion is consistentwith that of Gordon (1984), although an-other aspect of these results contrast withher findings. Apart from the slight polishingof enamel which resulted from tumbling aspecimen with quartz pebbles, the mostsignificant alteration to dental microwearoccurred after abrasion by the smallest (me-dium-sized sand) sediment particles, in con-trast with Gordon (1984), who found thatthe degree of modification to the dentalmicrowear was positively correlated withthe size of the abrasive particle. However,Gordon does not give enough detail about

372 T. KING ET AL.

the sizes of sediments she used in her experi-ments for any discussion of the reasons whyher results contrast with those in the pres-ent study.

SUMMARYThe experiments conducted here indicate

that dental microwear patterns tend to beobliterated rather than secondarily modifiedby taphonomic processes. These results arereassuring in that taphonomically altereddental microwear patterns can be readilyidentified. With the careful examination offossil material, dietary inferences are notlikely to be clouded by the effects of postmor-tem processes. Exposure to acids producedextensive erosion of the enamel and mi-crowear, with lower pH producing a greatereffect, and several Pasalar hominoid speci-mens display this type of pattern of alter-ation of the dental microwear. No alterationto the dental microwear occurred when aspecimen was exposed to an aqueous solu-tion of carbonatite ash. Polishing of theenamel and removal of finer dental mi-crowear features occurred when a specimenwas tumbled with the largest sediment par-ticle size used in these experiments. Noalteration to the microwear patterns wasobserved when a specimen was abradedwith the medium-sized sediment particles.The greatest modification to dental mi-crowear features occurred after tumblingwith the smallest sediment particles wherecomplete obliteration of microwear featureswas observed. Two specimens from Pasalardisplay this type of modification.

ACKNOWLEDGMENTSWe are grateful to Prof. Berna Alpagut for

access to the Pasalar specimens. We thankDr. Yolanda Fernandez-Jalvo, Prof. MikaelFortelius, and Mr. Christophe Soligo forseveral helpful discussions and for theircomments on the text. We are also gratefulto two anonymous reviewers for their com-ments on the text. This work was madepossible by a scholarship from The NaturalHistory Museum to T.K. and awards fromThe Central Research Fund (University of

London) and The Calouste Gulbenkian Foun-dation to T.K.

LITERATURE CITEDAlpagut B. 1990. A short history of the excavations at

the Miocene site at Pasalar, Turkey. J Hum Evol19:337341.

Alpagut B, Andrews P, Martin L. 1990. New hominoidspecimens from the Middle Miocene site at Pasalar,Turkey. J Hum Evol 19:397422.

Andrews P. 1990. Palaeoecology of the Miocene faunafrom Pasalar, Turkey. J Hum Evol 19:569582.

Andrews P. 1995. Time resolution of the Miocene faunafrom Pasalar. J Hum Evol 28:343358.

Andrews P, Alpagut B. 1990. Description of the fossilifer-ous units at Pasalar, Turkey. J Hum Evol 19:343361.

Andrews P, Tobien H. 1977. New Miocene locality inTurkey with evidence on the origins of Ramapithecusand Sivapithecus. Nature 268:699701.

Alpagut B, Andrews P, & Martin L. 1990. New hominoidspecimens from the Middle Miocene site at Pasalar,Turkey. J Hum Evol 19:397422.

Bozzola JJ, Russell LD. 1992. Electron microscopy.Boston: Jones and Bartlett Publishers.

Gordon KD. 1980. Dental attrition in the chimpanzee.Ph.D. dissertation, Yale University.

Gordon KD. 1982. A study of microwear on chimpanzeemolars: implications for dental microwear analysis.Am J Phys Anthropol 59:195215.

Gordon KD. 1983. Taphonomy of dental microwear: canfossil microwear be studied productively? Am J PhysAnthropol 60:200.

Gordon KD. 1984. Taphonomy of dental microwear, II.Am J Phys Anthropol 64:164165.

Gordon KD. 1988. A review of methodology and quantifi-cation in dental microwear analysis. Scanning Mi-crosc 2:11391147.

Kay RF, Hiiemae KM. 1974. Jaw movement and toothuse in recent and fossil primates. Am J Phys Anthro-pol 40:227256.

Keith A. 1924. Neanderthal man in Malta. J RoyalAnthropol Inst 54:251275.

King T, Aiello LC, Andrews P. 1999. Dental microwear ofGriphopithecus alpani. J Hum Evol 36:331.

King TC. 1997. Dental microwear and diet in Griphopi-thecus alpani. Ph.D. dissertation, University of Lon-don.

King TC, Aiello LC, Andrews P. 1994. Dental microwearand diet of Griphopithecus alpani. Am J Phys Anthro-pol Suppl 18:124.

Puech P-F, Prone A, Roth H, Cianfarani F. 1985. Repro-duction experiementale de processus dusure des sur-faces dentaires des Hominides fossiles: consequencesmorphoscopiques et exoscopiques avec application alHominide I de Garusi. C R Acad Sci III 301:5964.

Taylor PD. 1986. Scanning electron microscopy of un-coated fossils. Palaeontology 29:685690.

Teaford MF. 1988. A review of dental microwear and dietin modern mammals. Scanning Microsc 2:11491166.

Teaford MF. 1991. Dental microwear: What can it tell usabout diet and dental function? In: Else J, Lee P,editors. Advances in dental anthropology. New York:WileyLiss Inc. 341356.

Teaford MF. 1994. Dental microwear and dental func-tion. Evol Anthrop 3:1730.

Ungar PS. 1994. Incisor microwear of Sumatran anthro-poid primates. Am J Phys Anthropol 94:339363.


THE SITETHE PROBLEMFig. 1.

MATERIALS AND METHODSRESULTSFig. 2.Fig. 3.Fig. 4.Fig. 5.Fig. 6.Fig. 7.Fig. 8.TABLE 1.TABLE 2.

DISCUSSIONSUMMARYACKNOWLEDGMENTSLITERATURE CITED

taphonomic processes on dental microwear

Documents

wearmarks dental microwear

effect ondental microwear

hominoid dental specimens

effect of taphonomic

finer microwear features

dental microweartania

london sw7

london wc1e6bt