This article was downloaded by: [UQ Library]On: 03 November 2014, At: 16:11Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Historical Biology: An InternationalJournal of PaleobiologyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/ghbi20

Molluscs, isotopes and ecologicalfractionation: Examples from theupper Palaeogene of the HampshireBasinS.A.H. Young aa Institut Royal des Sciences Naturelles de Belgique , RueVautier 29, Bruxelles, BelgiumPublished online: 10 Jan 2009.

To cite this article: S.A.H. Young (1992) Molluscs, isotopes and ecological fractionation:Examples from the upper Palaeogene of the Hampshire Basin, Historical Biology: AnInternational Journal of Paleobiology, 6:4, 251-261, DOI: 10.1080/10292389209380434

To link to this article: http://dx.doi.org/10.1080/10292389209380434

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoeveras to the accuracy, completeness, or suitability for any purpose of the Content. Anyopinions and views expressed in this publication are the opinions and views of theauthors, and are not the views of or endorsed by Taylor & Francis. The accuracyof the Content should not be relied upon and should be independently verifiedwith primary sources of information. Taylor and Francis shall not be liable for anylosses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connectionwith, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.

Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

Historical Biology, 1992, Vol. 6, pp. 251-261 © 1992 Harwood Academic Publishers, GmbHReprints available directly from the publisher Printed in the United States of AmericaPhotocopying permitted by license only

MOLLUSCS, ISOTOPES AND ECOLOGICALFRACTIONATION: EXAMPLES FROM THE UPPER

PALAEOGENE OF THE HAMPSHIRE BASIN

S.A.H. YOUNG

Institut Royal des Sciences Naturelles de Belgique, Rue Vautier 29,B-1040 Bruxelles, Belgium

(Received April 16,1992)

Carbon and oxygen isotopic analyses from upper Palaeogene molluscs collected in the Hampshire Basin(S. England) show that, in addition to long-term trends in the data caused by climatic change, there isvariation within samples collected from any one horizon. This variation is not attributable to diagenesis orother "noise". Linear trends in data from the meso- and oligohaline organisms are salinity-dependent, as isa differentiation into clusters of the marine and brackish water animals. Within the marine organisms thereis further distinction between taxa, controlled by micro-environment. This is quite distinct fromdisequilibrium precipitation, as shell growth occurs in equilibrium with local isotopic ratios, though thismay not necessarily be the same as contemporary mean ocean values. This "ecological fractionation" canseriously affect the isotopic signal from a suite of fossils. Its existence should therefore be borne in mindwhen interpreting any biogenic isotope data from diverse taxa, localities, or micro-habitats.

KEY WORDS: Carbon and oxygen isotopes, fractionation, molluscan palaeoecology, salinity,Palaeogene.

INTRODUCTION

The widespread dissemination of DSDP data now means that many Earthscientists are accustomed to handling oxygen and carbon stable isotope dataderived from oceanic foraminiferal samples. Initial over-simplifications ininterpretation have largely been recognized and there is now a more sophisticatedappreciation of the natural phenomena which leads to variability and trends withindata (e.g. Shackleton, 1986; Shackleton and Boersma, 1981; Kennett and Barker,1990).

It is now widely recognized that when interpreting oxygen and carbon isotopesignals from foraminiferal samples, allowance has to be made for the ecologicalpreferences of the taxa being discussed (water depth, salinity, vertical migrationetc.). In short, to interpret the data correctly, one must be constantly aware of theenvironment in which the organisms lived and their autecology.

To date similar consideration of ecological effects on molluscs has proveddifficult. This is hardly surprising when one considers that the potential number ofecological variables which a shallow-water dwelling mollusc might be exposed tois far greater than that of a foraminifera in the relatively stable ocean environment.

During a recent study of the isotopic signals from upper Palaeogene molluscs ofthe Hampshire Basin (Young, 1992) observed variation in the data was attributableto two sources. The chief signal was a long-term trend through time caused bydeclining temperature in the Late Eocene and Early Oligocene (the subject of a

251

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

252 S.A.H. YOUNG

later manuscript). However the data also displayed significant second ordervariation within each horizon (see Figures 1 and 2).

Was this just "noise", an indication of diagenetic alteration, or a genuineenvironmental signal? This is the question addressed by this paper.

MATERIALS AND METHODS

The material for study came from the upper Palaeogene of the Isle of Wight andthe adjacent mainland in southern England. The sampled sequence starts in theupper Ypresian (Early Eocene) and extends up into the Rupelian (EarlyOligocene). The sediments are highly varied, representing a wide range ofsedimentary environments and facies, ranging from normal marine, throughlagoonal (both hyper- and hyposaline) and estuarine to fluvial and lacustrine.Together they comprise the infill of the Hampshire Basin which in turn is part ofthe larger Northwest European Anglo-Paris-Belgian Basin.

A variety of bivalve and gastropod taxa was chosen, with both calcitic andaragonitic shells. Most were from marine or near-marine facies, though somebrackish and freshwater organisms were also sampled. All specimens togetherwith their ages (as interpolated from the Berggren et al., 1985 time scale) arelisted in Table 1.

The selected shells were thoroughly cleaned of adhering material, washed in anultrasonic bath and either crushed in an agate pestle and mortar or sampled with asmall hobby drill. Great care was taken to ensure that the powder was free of anyfragments or chips.

Samples were not roasted, as pyrite was negligible or non-existent. Nor were

i ' r30 40 50

AGE IN Ma.Figure 1 Age distribution of ^ ' O data.

60

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

MOLLUSCS, ISOTOPES AND ECOLOGICAL FRACTIONATION 253

samples bleached. Land (1989) reviews the various methods used to treat samplesprior to analysis and concludes that all the commonly used techniques have adifferent effect on the shell. "Organic material" is blamed but no clear explanationhas been found to demonstrate why different isotopic values are derived afterdifferent preparations. Samples were therefore not pretreated since the processesinvolved appear to significantly alter the carbonate in an unpredictable way.

Between 250 and 500 fig sub-samples of powder were placed in stainless steelbuckets. Smaller samples would have required the use of the "cold finger"cryogenic condenser, resulting in less accurate analyses. With larger samples, orsamples containing uncrushed fragments, there would have been a danger that theprevious sample might still have been effervescing when the next sample dropped.

Up to 40 samples at a time were loaded into an automatic sample changer. Twostandards were loaded at the beginning and end of each run and one in the middleto ensure that any "drift" could be corrected for. The samples were thensequentially dropped into a common acid bath (105% H3PO4 held at 90°C by awater jacket) to digest. The evolved CO2 was drawn into a VG Isotech massspectrometer and the oxygen and carbon isotopes analysed. Several samples of gaswere taken from each powder and compared with a reference gas. These data werethen fed to a computer which determined if further gas samples were needed (thiswas the case if the reproducibility of analyses from a powder fell below a presetlimit). A mean value was then calculated for each sample, corrections applied forthe digestion having occurred at high temperature and for any "drift", and internalconfidence limits applied. The entire process was fully automated, a complete setof 40 samples taking approximately 24 hours to run.

30i ' r40 50

AGE IN Ma.

Figure 2 Age distribution of #3C data.

60

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

254 S.A.H. YOUNG

Table 1 Stable isotope sample data for molluscs from the Upper Palaeogene of the Hampshire Basin, givingsample number, taoxonomy, age in Ma. as interpolated from the Berggren el al. (1985) time scale andcorrected values of <5"C and 8*O.

SAMPLE TAXONOMY5455566263676076827087845317665981681642162173677691

464724252627283132404142433033

16116220421392484950513435

211212167657879907

8889

9397

CarditaCarditaCarditaCarditaCarditaCarditaCarditaCarditaCarditaCarditaCarditaCarditaCrassatellaCrassatellaCrassatellaCrassatellaCrassatellaCrassatellaGlycmerisGlycymerisGlycymerisNemocardiumNemocardiumNemocardiumVerier icardiumVerier icardiumVerier icardiumVenericardiumVerier icardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumVenericardiumCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaCubitostreaOytreaOstreaOstreaOstreaOstreaOstreaOstrea

Aqe Ma42.742.442.442.442.442.442.041 .641 .541 .440.540.043.042.442.442.041 .541 .448.045.045.049.541.641 .454.054.054.051 .051 .051 .051 .051 .050.550.550.050.050.050.049.549.548.048.046.543.538.550.050.050.050.049.549.546.046.044.542.441 .541 .538.538.238.037.237.237.237.2

.6 13C0.701.141 .711 .771 .871 .300.001 .360.980.821 .011 .081 .800.981 .920.681 .671 .511 .653.471 .802.160.780.97

-0.81-0.90-0.782.131 .571 .482.000.931 .211 .760.141 .301 .39

-0.030.440.851 .331 .181 .201 .121 .98

-2.77-3.81-2.71-3.121 .641 .073.32

-0.241 .761 .221 .661 .882.19

-2.32-2.62-0.93-0.95-1 .31-1 .24

6 180-2.70-3.36-3.61-3.13-2.77-3.04-0.39-2.64-3.13-2.50-2.2.2-2.56-3.52-2.36-2.36-1 .75-2.84-2.24-3.38-1 .40-2.15-1 .48-2.90-2.39-4.48-4.67-4.17-4.15-4.89-4.48-4.96-3.52-3.95-4.96-5.29-4.34-4.36-5.24-5.05-3.74-2.71-2.95-2.80-3.71-2.47-4.74-5.05-4.73-4.76-3.40-3.20-1 .90-1 .68-3.28-3.09-3.30-2.68-3.14-2.24-2.47-2.34-2.37-2.19-1 .73

SAMPLE TAXONOMY Aae Ma.1094959612

1001012398993

52293839444452037

1631651662052062092102181681692142155761645874758071728918

10220720856151613155151158160150157153159149154152

PolymesodaPolymesodaPolymesodaCorbiculaPolymesodaPolymesodaPolymesodaPolymesodaPolymesodaPolymesodaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaTurritellaVolutaAthletaClavilithesClavilithesPseudocominelPseudocominelPirenellaPirenellaPirenellaMelanoidesMelanoidesMelanoidesMelanoidesMelanopsisMelanopsisViviparusViviparusGalbaGalbaGalba

37.037.037.037.035.035.035.033.033.033.053.553.551.050.550.550.050.050.049.549.548.047.547.546.546.546.046.045.044.544.543.543.542.442.442.442.041 .641 .641 .541 .041 .038.542.432.746.046.038.338.333.033.032.738.537.036.733.037.037.038.634.539.038.438.1

6 13C-1 .37-0.74-2.59-2.45-3.76-3.24-3.32-1 .54-3.06-2.890.441 .082.902.172.312.631 .872.801 .88

-0.031 .711 .512.122.332.571 .553.362.253.363.831 .332.793.933.322.772.781 .951 .623.452.262.460.140.03

-2.030.801 .16

-1 .20-2.54-3.78-2.39-2.64-2.23-0.72-1 .79-1 .450.04

-1 .57-4.88-2.09-4.57-4.602.86

6 180-2.39-2.04-1 .89-2.19-2.85-2.60-2.66-1.36-1.93-2.24-4.25-4.85-5.38-3.58-4.64-2.43-3.68-3.91-4.47-2.39-2.33-2.55-2.56-3.02-2.90-0.69-1.21-1 .06-3.41-2.69-4.21-3.63-2.64-2.90-2.09-2.10-2.28-2.85-1 .99-3.94-2.66-5.13-3.03-0.69-2.94-1 .64-2.03-1 .57-2.28-1 .92-1 .980.05

-1 .42-1 .57-0.96-1 .63-2.260.92

-0.06-1 .822.821 .66

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

MOLLUSCS, ISOTOPES AND ECOLOGICAL FRACTIONATION 255

RESULTS

The results of the isotopic analyses are presented in Table 1. Oxygen and carbonare reported in the conventional £ notation (Faure, 1977), i.e.

<S=(Rx-Rstd)/RstdxlO3

where Rx is the ratio of I3C/12C or 18O/16O in the sample for carbon and oxygenrespectively and Rstd is the ratio of those isotopes in a standard (PDB).

All the data lie within the values given by Hudson (1977) for common marinelimestones (even the fresh-water shells, due to vital effects discussed below). Thecarbon values are also in broad agreement with those of modern sediments(Hudson, 1977) and modern near surface sea water (Morrison and Brand, 1986)indicated by the box in Figure 3. However the oxygen data are generally muchlighter than modern values due to the warmer climate of the lower Palaeogene(Young, 1992).

INTERPRETATION

Figures 3 and 4 show the data from the marine molluscs and those from thebrackish and freshwater material respectively. As can be seen these data are highlyscattered. Part of this scatter can be explained by declining temperature andchanging climate (Young, 1992 and in prep). The difference in equilibrium values

5 —|

& —

3 —

2 -

-1 —

0 —o

So -1 -

o

-

-3 —

-4 —

-o

-6

0 0

o o o0

0 0 0 »0 fen J

o a

00 •

1 ' 1 ' 1

-5 -4 -3

c

. 00°

/?«d s

• o

ta

1 1 1 1

-2 -1

8

i Modern; surfacei water

o :

• ..:

«

, | , |

0 1

nearsea

I l

2

o TURRITELLA• NEOGASTROPODE CRASSATELLA• NEMOCARDIUMB GLYCYMERIS• CARDIUM

3

Figure 3 8iClB%O plot for marine molluscs from the Upper Palaeogene of the Hampshire Basin.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

256 S.A.H. YOUNG

o —

4 —

3 —

2 -

0 -o- -1 —

-2 -

-3 -

-4 -

c-0

El

^ D D n S ^

D

A

® E •

i | i | , | , | , | , | ,

B CUBITOSTREA• OSTREAD VENERICARDIUM

H CORBICULA• PIRENELLA» MELANOIDES

* FW GASTRAPODS

-6 -5 -4 -3 -2 -1

5180

Figure 4 5"C/d18O plot for brackish and freshwater molluscs from the Upper Palaeogene for the HampshireBasin.

of aragonitic and calcitic shell material (e.g. Anderson and Arthur, 1983;Grossman and Ku, 1986) appears to have had little effect as there is no cleardistinction in Figure 4 between the chiefly calcitic organisms (i.e. Cubitostrea andOstrea) and the other, aragonitic, molluscs. What controlled the rest of thisvariation? Diagenesis? Disequilibrium?

Tests for Possible Diagenesis

The most likely cause of scatter in such a data set is diagenesis. Post-burialalteration of shell material would result in the original signal being overprinted,particularly the oxygen signal, which is more susceptible to change than carbon.Assessment of the geochemical preservation of the shells was thereforeundertaken using the following criteria—

Macrosopic Appearance

Only shells which appeared to be well preserved were selected; chalky or pyritizedmaterial was excluded where possible. Similarly an effort was made not to takesamples from areas of shell which had been subjected to attack by boring clionidsponges, as this might have isotopically altered the carbonate.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

MOLLUSCS, ISOTOPES AND ECOLOGICAL FRACTIONATION 257

Table 2 ICP sample data for a subset of samples from the Upper Palaeogene of the Hampshire Basin givingsample number and Sr, Mg, Mn and Na concentrations.

412263638455458638084859599102137

pm3421.831607.504689.662057.943762.683650.154294.542596.635277.492803.262403.562023.061964.811838.172514.791571.97

.89

.12

Mq-ppm85.5582.84258.13111 .151.149.65209.0674.33

208.38109.07269.0476.5262.5642.2267.06113.64

Mn-ppm33.4344.3836.4529.9769.9851 .5742.2228,7446.0735.6865.2838.7841.0629.9075.9426.52

Ha-ppm2369.712159.762482.762877.122241.382396.362306.902685.832743.462833.842482.693123.691867.063482.853214.993570.08

Microscopic Examination

Several shells were fractured, mounted on aluminium stubs, spatter coated withgold and examined using SEM. The growth patterns and microstructure of thesamples examined were indistinguishable from those of Recent shell material.

XRD Analysis

A representative set of 17 sub-samples from material used in the stable isotopeanalysis was made into slurries with acetone, mounted on glass slides andanalysed by X-ray diffractometer. There was no evidence that significant aragonitehad converted to calcite, indicating that mineralogically the material wasunaltered.

Cation Geochemistry

Sixteen of the shell samples were selected, dissolved in acid and subjected toanalysis by ICP emission spectroscopy (see Table 2 for results). The chosensamples included those which appeared most likely to have been altered. Mg, Sr,Mn and Na were analysed and the values compared with those found in Recentmaterial. The expected ranges quoted in Morrison and Brand (1986) for modernaragonitic shell carbonate are Mg 102-104 ppm, Sr 102-104 ppm, Na 103-6x103 ppm and Mn 0-103 ppm. Any substantial deviation from this range issuggestive of diagenetic alteration. The values found in the samples all lie withinthe expected range.

Mg/Mn ratios may also be used as an indicator of chemical alteration (Brandand Morrison, 1987). The results obtained in this study all fall well within therange found in modern, organically precipitated aragonite.

Clearly the samples are chemically all very well preserved. Four independenttests failed to find any evidence of post-mortem alteration. Therefore the scatterfound in the data is unlikely to be the effect of diagenesis and most probablyrepresents original signals from the environment in which the shells grew. Whatmight these signals represent?

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

258 S.A.H. YOUNG

-5 -45180

-3 -2

Figure 5 83C/8"'O plot for Cubitostrea from the Upper Palaeogene of the Hampshire Basin, showingsalinity dependent trend.

Intrageneric Variation

Figure 4 shows that within certain genera {Cubitostrea, Corbicula, Pirenella),there is a distinct trend in the 5*3C/18O plots. <S18O and <5'3C show a clear positiverelationship. This is best demonstrated in Cubitostrea (Figure 5) in which there isa correlation coefficient (r) of 0.939 if one rather rolled and corroded specimen(212) is excluded (r=0.674 if this is included).

This relationship is very similar to that observed by Mook (1971) whodemonstrated that in modern estuarine and marginal marine waters there was alinear relationship between £18O and &3C controlled by salinity. In each riversystem there was a trend of heavier isotopic values with increasing salinity. That itis salinity control which is determining the intrageneric variation is madeincreasingly likely by the fact that it is the brackish water taxa which display thetrend most clearly.

Intergeneric Variation

Some of this variation is clearly accounted for by the disparate salinity regimes inwhich the different taxa lived. Salinity (or rather the proportion of meteoric watermixing with sea water) has a well understood effect on oxygen and carbon isotopicratios, as demonstrated above for individual taxa. It is therefore no great surprisethat the brackish water organisms (Corbicula, Pirenella, Ostrea, etc.) plot outquite discretely from the marine shells. The mesohaline organisms (e.g.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

MOLLUSCS, ISOTOPES AND ECOLOGICAL FRACTIONATION 259

Veriericardium and Cubitostrea) occupy a position partially overlapping themarine cluster but also intermediate between the two groups. For instanceVenericardium has a mean carbon value of 0.93 per mil, only a little lower thanthat of Cardita (1.15). The genuinely freshwater samples (Melanopsis, Galba andViviparus) show anomalously heavy oxygen isotope ratios (mean—0.86). Thereason why is not fully understood and is the subject of current research; howeverit is likely to be either the consequence of isotopic fractionation in shallowevaporating lakes and ponds, or a product of the respiratory physiology of theorganisms (e.g. the source from which they derived their oxygen—the air or thewater). These processes would appear to satisfactorily explain the observedvariation in oxygen isotope signatures of the different groups.

The most novel and surprising aspect of the results is however still unexplained,i.e. the degree of variation between the different carbon isotopes in the marineorganisms. Three distinct clusters may be observed, each significantly differentfrom the other (at /)<0.002) (see Figure 3). The turritellids have the heaviestvalues (mean <5I3C=2.23), the bivalves (Crassatella, Nemocardium andGlycymeris) have an intermediate value (mean=1.15), while the neogastropods("Voluta", Clavilithes and Pseudocominella) have very light values (mean=-0.63). These samples come from similar facies and in some cases the same bed,so might this represent disequilibrium precipitation of some of the shell material?

Several organisms, particularly those which are either genuinely orcommensally photosynthetic, are known to fractionate isotopes metabolically.These "vital effects" can have marked influence on the delta values of thecarbonate since relatively light or depleted respiratory CO2 becomes incorporatedinto the carbonate resulting in deposition of shell that is not isotopically identicalwith the environment in which it grew. However molluscs (with the exception ofcertain bivalves such as the giant clam Tridacna which have symbioticzooxanthellae (Jones et al., 1986)) are known not to fractionate extensively(Epstein et al., 1953); though they do show seasonal variation which might reflectmetabolic processes such as growth rate and spawning (i.e. vital effects) inaddition to fluctuations in the isotopic value of the water (due to e.g. temperature)(Wefer and Berger, 1991). The variations discussed here are however averagedthrough time by the sampling process and are therefore not of this nature.Similarly kinetic effects cannot be invoked as there is no evidence to suggest thatdeposition of carbonate occurred at different rates between the different groups.("Kinetic effects" are brought about by differences in precipitation-, or rathercrystal growth-, rate of the carbonate. Turner (1982) discusses this for inorganicprecipitation, while the review of Wefer and Berger (1991) suggests the possibilityof such processes in the deposition of molluscan shells).

Diagenetic alteration and salinity variation have also been discounted aspotential controls; therefore it must be some other factor, intrinsic to the ecologyof these organisms, which controlled this second-order variation.

A combination of diet and habitat seem likely to have been responsible for an"ecological fractionation" of the isotopes within these organisms. Theneogastropods are all members of benthonic carnivorous groups. They would havebeen in contact with clean, well oxygenated seawater at all times. The bivalvessampled were all shallow infaunal filter feeders, and would thus have beenexposed to slightly less well oxygenated conditions, and water already affected bythe metabolism of other organisms. The turritellids are all infaunal ciliary filterfeeders living in an isotopically heavy, fractionated, micro-environment. Decayingdetritus in the sediment, already partially digested by bacteria and other

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

260 S.A.H. YOUNG

organisms, is depleted in the lighter carbon isotope, leaving a heavy residue onwhich the turritellids feed. Similarly CO2 (as dissolved HCO3~) is metabolicallydepleted in the lighter isotope. These depleted sinks are the source from which theorganisms derived their carbonate for shell construction. Consequentially theresultant carbonate is isotopically very heavy.

CONCLUSIONS

The above data clearly show that great care is needed when sampling molluscanpopulations for stable isotope analysis, and in the interpretation of the results. Inaddition to variation caused by salinity differences, quite subtle distinctions inhabitat can apparently lead to marked isotopic differentiation. This is quite distinctfrom disequilibrium, i.e. deposition of carbonate out of equilibrium with theenvironment in which it was deposited. In the case of ecological fractionationdeposition is in equilibrium with the surrounding micro-environment, though thisvalue may differ markedly from that of a notional oceanic mean. The biology ofthe organisms being sampled must constantly be borne in mind. The existence ofthis ecological fractionation adds yet another constraint to the interpretation ofoxygen and carbon isotopic data.

Acknowledgements

Dr D. Mattey kindly ran the isotope analyses at Royal Holloway and Bedford New College, for which Iam extremely grateful. Mr J. Todd generously donated some of the molluscan material from the top ofthe Bracklesham Group. I would also like to thank my supervisor, Prof A. Hallam and an anonymousreviewer for critically reading earlier draughts of the manuscript and making several helpful comments.This work was undertaken as part of the author's NERC sponsored Ph.D. The manuscript was preparedfor publication while in receipt of a Royal Society European Exchange Fellowship. Both sources offunding are gratefully acknowledged.

References

Anderson, T.F. and Arthur, M.A. (1983) Stable isotopes of oxygen and carbon and their application tosedimentalogic and palaeoenvironmental problems. SEPM short course, 10, Stable isotopes insedimentary geology, 1-151

Berggren, W.A., Kent, D.V. and Flynn, J.J. (1985) Palaeogene geochronology and chronostratigraphy.Memoir of the Geological Society, 10, 141-186

Brand, U. and Morrison, J.O. (1987) Biogeochemistry of marine fossil invertebrates. GeoscienceCanada, 14, 85-107

Epstein, S., Buchsbaum, R., Lowenstam, H.A. and Urey, H.C. (1953) Revised carbonate-water isotopictemperature scale. Geological Society of America Bulletin, 64, 1315-1326

Faure, G. (1977) Principals of Isotope Geology. John Wiley and Sons, New York, 464 pp.Grossman, E.L. and Ku, T-L. (1986). Oxygen and carbon isotope fractionation in biogenic aragonite:

temperature effects. Chemical Geology (Isotope Geosciences Section), 59,59-74Hudson, J.D. (1977) Stable isotopes and limestone lithification. Journal of the Geological Society of

London, 133,637-660Jones, D.S., Williams, D.F. and Romanek, C.S. (1986) Life-history of symbiont-bearing giant clams

from stable isotope profiles. Science 231,46-48Kennett, J.P. and Barker, P.F. (1990) Latest Cretaceous to Cenozoic climate and oceanographic

developments in the Weddall Sea, Antarctica: An ocean-drilling perspective. Proceedings of theOcean Drilling Program, 113,937-960

Land, L.S. (1989) The carbon and oxygen isotopic chemistry of surficial Holocene shallow watercarbonate sediment and Quaternary limestone and dolomite. In Handbook of environmental isotopic

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014

MOLLUSCS, ISOTOPES AND ECOLOGICAL FRACTIONATION 261

geochemistry,Vol. 3, The marine environment. A, edited by P. Fritz and J. Ch. Fortes, pp. 191-217,Elsevier

Mook, W.G. (1971) Palaeotemperatures and chlorinities from stable carbon and oxygen isotopes of shellcarbonate. Palaeogeography, Palaeoclimatology, Palaeoecology, 9, 245-263.

Morrison, J.O. and Brand, U. (1986) Geochemistry of recent marine invertebrates. Geoscience Canada,13,237-254

Shackleton, N. (1986) Palaeogene stable isotope events. Palaeogeography, Palaeoclimatology,Palaeoecology, 57,91-101

Shackleton, N. and Boersma, A. (1981) The climate of the Eocene ocean. Journal of the GeologicalSociety of London, 138, 153-157

Turner, J.G. (1982) Kinetic fractionation of carbon-13 during calcium carbonate precipitation.Geochemica et Cosmochemica Acta, 46, 1183-1193

Wefer, G. and Berger, W.A. (1991) Isotope paleontology: growth and composition of extant calcareousspecies. Marine Geology 100,207-248

Young, S.A.H. (1992) Unpublished Ph.D. thesis "Environmental changes and molluscs in the upperPalaeogene of North-West Europe". University of Birmingham. 326 pp.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

16:

11 0

3 N

ovem

ber

2014