ELSEVIER Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

Palaeoecology of the biserial planktonic foraminifer Heterohelixmoremani (Cushman) in the late Albian to middle Turonian

Circum-North Atlantic

Alexandra J. Nederbragt a,Ł, Robert N. Erlich b, Bruce W. Fouke a,1, Gerald M. Ganssen a

a Instituut voor Aardwetenschappen, Vrije Universiteit, de Boelelaan 1085, 1081 HV Amsterdam, Netherlandsb Amoco Production Company, 501 WestLake Park Boulevard, Houston, TX 77253, USA

Received 5 May 1997; accepted 23 April 1998

Abstract

The Cretaceous Heterohelix moremani (Cushman) was the only biserial planktonic foraminiferal species from its firstappearance in the late Albian up to the Cenomanian=Turonian boundary. Within that time, it increased gradually inabundance relative to other planktonic foraminifera in five Circum-North Atlantic sections. It is generally rare in upperAlbian sediments, common in most of the Cenomanian and very abundant in sediments representing the latest CenomanianOceanic Anoxic Event. Short-term variations on the overall abundance trend correlate with positive excursions in thebulk carbonate Ž13C record. Maximum rain rates of H. moremani during OAE2 show that this species was an opportunistthat did well in extreme conditions, but its overall distribution indicates that it is not necessarily a marker for very highpalaeoproductivity environments. Stable oxygen and carbon isotope measurements on foraminiferal species indicate that H.moremani was a surface water dweller at least in part of its geographic range, but incorporated 13C out of equilibrium withambient seawater. It is depleted in Ž13C relative to other planktonic foraminifera, which is attributed to vital effects relatedto its opportunistic character. 1998 Elsevier Science B.V. All rights reserved.

Keywords: palaeoecology; middle Cretaceous; Heterohelicidae; planktonic foraminifera; carbon isotope stratigraphy;Circum-North Atlantic

1. Introduction

Amongst diverse genera of planktonic foramini-fera are a number of groups of small sized oppor-tunists with long biostratigraphic ranges, but with

Ł Corresponding author. Present address: Department of Geo-logical Sciences, University College London, Gower Street,London WC1E 6BT, UK. Fax: C44 (171) 388-7614; E-mail:A.Nederbragt@ucl.ac.uk1 Present address: Department of Geology, University of Illinois,1301 West Green Street, Urbana, IL 61801, USA.

a patchy distribution in time and space. Mid-Creta-ceous to Recent triserial forms (Guembelitria Cush-man and Gallitellia Cushman), mid-Cretaceous andCenozoic biserial species (Heterohelix moremani(Cushman) and Chiloguembelina spp. and Strepto-chilus spp.) and enroled biserial forms (Cassige-rinella Pokorny and Cassigerinelloita Stork) are oflimited use for biostratigraphy, because of theirlong ranges. In part, their stratigraphic distribu-tion is imperfectly documented, because their smallsize means that they are easily overlooked (Resig

0031-0182/98/$19.00 c 1998 Elsevier Science B.V. All rights reserved.PII: S 0 0 3 1 - 0 1 8 2 ( 9 8 ) 0 0 0 8 9 - 3

116 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

and Kroopnick, 1983; Kroon and Nederbragt, 1990;Koutsoukos, 1994). However, their geographic dis-tribution gives important palaeoceanographic infor-mation. Abundant occurrences of these forms havebeen related to extreme or variable surface waterconditions as in shelf seas or in upwelling areas(Leckie, 1987; Kroon and Nederbragt, 1990; Li andRadford, 1992); or were interpreted as indicative forthe presence of an intense Oxygen Minimum Zone(Resig and Kroopnick, 1983; Boersma and PremoliSilva, 1989).

The biserial planktonic foraminifer Heterohelixmoremani (Cushman) is an exception in the list ofopportunists to the extent that it is the only formto evolve, forming the ancestral stock of diverseLate Cretaceous heterohelicid genera. This smallbiserial species ranged into the Santonian (Neder-bragt, 1991) but from its first appearance in the lateAlbian until the Cenomanian–Turonian boundary itformed a monospecific genus. Heterohelicids startedto diversify during the latest Cenomanian OceanicAnoxic Event (OAE2). Heterohelix moremani wasinterpreted as an opportunist, with geographic varia-tion in abundance giving palaeoenvironmental infor-mation. Leckie (1987) concluded that H. moremani isindicative for epicontinental sea deposits; Nederbragt(1991) thought that it was a marker for variable con-ditions more in general. Yet regional palaeoenviron-mental conditions may not be the only control on thedistribution of this species. Variation in abundance ofH. moremani was found to allow for biostratigraphicsubdivision of upper Albian to lower Turonian sedi-ments in Venezuela. Keeled biostratigraphic markerspecies were scarce in these sediments because ofextreme surface water conditions. The presence ofsuccessively rare, common and abundant biserialplanktonic foraminifera was used in an unpublishedbiostratigraphic study to distinguish, respectively,upper Albian, lower and middle Cenomanian andupper Cenomanian or younger sediments.

The aim of this study is to document the ge-ographic and stratigraphic distribution of mid-Cre-taceous biserial planktonic foraminifera in order toevaluate their significance for biostratigraphy and(or) palaeoceanography. If H. moremani is primarilya palaeoenvironmental indicator then any biostrati-graphic zonation using this form, as was done for theVenezuelan sections, is at best regional and biostrati-

graphic boundaries are potentially diachronous. Con-versely, its palaeoceanographic usefulness would belimited if variation in abundance is synchronous overa wide area covering different environments. For thisstudy, distribution patterns of late Albian to earlyTuronian heterohelicids were documented quantita-tively in Circum-Atlantic sections that can be fittedin a standard planktonic foraminiferal zonation. Inaddition, stable oxygen and carbon isotopes weremeasured on H. moremani and other foraminiferalspecies, to establish their position in the water col-umn, and to compare their ecology to triserial andCenozoic biserial planktonic foraminifera.

2. Material and methods

Distribution patterns of H. moremani were com-piled from washed residues from marl samples aswell as thin-sections from limestones from four landsections and one Deep Sea Drilling Project (DSDP)section (Fig. 1). Samples from two more DSDP siteswere added for stable isotopes measurements on in-dividual species. To obtain washed residues, marlysamples were dried overnight at 50ºC, soaked inwater, and rinsed over a 45 µm sieve preferably,as only a small part of a H. moremani populationis large enough to be retained in a 63 µm sievescreen. Samples from all three DSDP sites and frompart of the Monjas section (Spain) consist of weaklyconsolidated sediments that could be washed overa 45 µm sieve. However, marly samples from theMellegue section (Tunisia) and some samples fromthe Monjas section are more consolidated and werewashed over a 63 µm sieve, as residues from a45 µm sieve were not well enough preserved forquantitative analysis. All washed residues were splitwith a microsplitter until a quantity remained thatcould be evenly distributed on a tray, from which300 biogenic grains or more were counted to es-timate relative frequencies of the main microfos-sil groups in the washed residues. These groupsinclude planktonic and benthic foraminifera, calci-spheres and radiolaria. The number of fragments ofplanktonic foraminifera were counted, and expressedas a percentage relative to whole tests, as an indicatorfor the extent of dissolution of foraminifera belowthe lysocline. Within the planktonic foraminiferal

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 117

Fig. 1. Palaeogeographic reconstruction of the mid-Cretaceous Circum-North Atlantic (after Barron et al., 1981 and Barron, 1987)showing sample localities. Numbers refer to DSDP Sites. Black dots indicate sections in which stratigraphic record of H. moremani wasstudied, open circles show sections from which additional samples were used for stable isotope measurements on species foraminiferalspecies.

assemblage, only biserial Heterohelix and triserialGuembelitria were counted separately from all otherplanktonic foraminifera.

Sections in Venezuela and near Cismon, NE Italy,as well as part of the Monjas and Mellegue sectionsconsist of limestones, from which thin-sections weremade. Numbers of foraminifera and their fragments,calcispheres and radiolaria per surface area wereestimated by counting all microfossils in subsequentfields of view, until at least 150 specimens werecounted. To estimate percentages of biserial versusspiral planktonic foraminifera, each thin-section wasfurther scanned until around 100 specimens werefound that could be positively identified as eitherbiserial or spirally coiled. Triserial planktonic formscannot be recognised consistently enough to givegood estimates of their abundance in thin-section.

All samples were scanned for the presence of bios-tratigraphic marker species, and are zoned followingthe zonation of Caron (1985) with minor modifica-tions. The Rotalipora subticinensis Zone and Rotal-ipora ticinensis Zone are combined into a single R.ticinensis–subticinensis Zone. The name Rotaliporaglobotruncanoides is used as a senior synonym of Ro-talipora brotzeni. The studied sections together coverthe upper Albian to middle Turonian. Heterohelixmoremani is the only biserial planktonic species un-til the Cenomanian=Turonian boundary. Heterohelix

globulosa (Ehrenberg), which has its first appearancenear this boundary, is frequent in Turonian samples.Heterohelix globulosa differs from H. moremani inhaving larger chambers, which increase more rapidlyin size through ontogeny, and in being covered withdistinct costae (Nederbragt, 1991). However, interme-diate forms are present, and the two species cannot bedistinguished consistently in thin-section. They aretherefore counted together, giving the abundance ofheterohelicids as a group relative to all other plank-tonic foraminifera.

Carbon isotope stratigraphy offers a tool for cor-relation in addition to biostratigraphic data, as thepattern of variation in stable carbon isotope compo-sition of pelagic carbonates during the mid-Creta-ceous is well documented. Three positive excursionshave been recorded in the upper Albian to Turo-nian (Jenkyns et al., 1994; Erbacher et al., 1996).A nearly 1‰ excursion in Ž13C occurred around theAlbian=Cenomanian (A=C) boundary, a brief 0.5‰Ž13C excursion is found spanning the CenomanianR. reicheli Zone, and a >2‰ Ž13C excursion marksthe start of the latest Cenomanian OAE2. For strati-graphic purposes, carbon and oxygen isotopes weremeasured on bulk sediments for limestone as wellas marly samples, to ensure compatibility of data fordifferent lithologies. Some 100 mg of powder wasdrilled from hard limestone samples; for less con-

118 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

solidated sediments approximately 0.5 g of materialwas ground in a mortar.

In addition, stable oxygen and carbon isotopemeasurements were done on individual planktonicforaminiferal species, to determine the position of H.moremani in the water column. Although morpho-logic preservation of foraminifera in washed residuesfrom the land sections and at DSDP Site 547 is good,specimens are calcite-filled and unlikely to give reli-able stable isotope data for individual species. There-fore, additional samples from DSDP Sites 137 and551, which contain air-filled foraminifera, were usedfor measurements on individual species. All isotopemeasurements were done on a Finnigan MAT 251mass spectrometer with an automated sample prepa-ration line. Samples were dissolved in 100% phos-phoric acid at 70ºC. Measurements are reported inthe standard δ notation relative to the PDB standard.Average values are shown for duplicate measure-ments. Analytical precision of an internal carbonatestandard, as well as of duplicated bulk sedimentsamples, is better than 0.10‰ (1¦ ) for oxygen and0.05‰ (1¦ ) for carbon.

3. Results

3.1. Chimana and Querecual formations, easternVenezuela

Deposition during the Albian–Turonian in east-ern Venezuela consisted mostly of outer shelf toupper slope sediments of the Chimana Formation,and upper slope sediments of the organic carbon-rich Querecual Formation (Hedberg and Pyre, 1944;Alberdi and Lafargue, 1993). The Chimana consistsof about 800 m of shales with thin sandstone andlimestone interbeds, and is apparently transitionaland conformable with the overlying 750 m thicksiliceous limestones and shales of the Querecual; thecontact between the two formations is diachronous(Direccion de Geologıa, 1970; Yoris, 1990).

Samples for this study were taken from fourlocations in eastern Venezuela (including the typelocalities), and represent a characteristic vertical se-quence through both units. The sections were studiedas part of an industrial project. Stratigraphic varia-tion in percentages of heterohelicids in the plank-

tonic foraminiferal assemblages is shown in Fig. 2.Age-assignments were tentative, based on scatteredoccurrences of Ticinella spp., which range to ap-proximately the top of the upper Albian, and of largeglobular Whiteinella specimens that are typical forthe latest Cenomanian OAE2 and younger Turoniansediments (Caron, 1985). Carbon isotope data do notoffer further means for correlation as original valueshave been altered (Fig. 2). Highly negative Ž18O val-ues indicate that diagenesis played an important role.In addition, Querecual samples have high organiccarbon content, and exchange with the light carbonisotopes of organic material may also have alteredthe Ž13Ccarbonate signature (Scholle and Arthur, 1980).

3.2. DSDP Site 547, off Morocco

A 350 m thick sequence of dark grey hemipelagicclays, in the lower part interbedded with mud brec-cias, was drilled at DSDP Site 547 (Hinz et al.,1984). Samples were taken in hemipelagic claysthroughout the section, avoiding mud breccias wherepresent. Leckie (1984) assigned the sediments to theupper Albian R. ticinensis Zone to middle Ceno-manian Rotalipora cushmani Zone. However, thepresence of R. cushmani in the upper part of thesection could not be confirmed in this study, and thetop of the section is considered to belong to thelower Cenomanian Rotalipora globotruncanoidesZone. An early Cenomanian age is supported bystable carbon isotope stratigraphy. The section showsan expanded positive excursion in Ž13C around theAlbian=Cenomanian boundary (Fig. 3), but the topof the section records at best only the base of the nextyounger, mid-Cenomanian excursion, which wouldbe expected to be present completely if the sectionextended into the R. cushmani Zone (Jenkyns et al.,1994).

Foraminifera are calcite-filled in almost all sam-ples, but faunas show good to excellent morphologicpreservation, with low degrees of fragmentation ofplanktonic foraminiferal tests (Fig. 3). Leckie (1984,1987) found that H. moremani is rare in the >63 µmsieve fraction. The species is much more abundantin the >45 µm sieve fraction, reaching frequenciesof more than 20%, but the stratigraphic pattern ofvariation is similar to that in the >63 µm fractionas analysed by Leckie (1987). Heterohelix more-

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 119

Fig. 2. Percentage of heterohelicids within the planktonic foraminiferal faunas in thin-sections from two eastern Venezuelan sections inthe Querecual and Chimana formations, with bulk carbonate Ž13C values. Note that samples are in stratigraphic order but that thickness isnot on a linear scale. Scattered, low Ž13C values in combination with Ž18O values ranging between �6 and �12‰ (not shown) indicatethat stable isotope values are altered.

mani is very rare in most of the upper Albian, butshows a first increase in abundance at the base ofthe A=C Ž13C excursion. Percentages are variablehigher in the section, but a second increase in over-all abundance occurs above the excursion. TriserialGuembelitria shows a more or less reversed pattern.It is abundant up to the top of the Ž13C excursion,and very rare above it. Within the excursion, highfrequencies of Guembelitria cenomana tend to cor-respond to low frequencies of H. moremani and viceversa. Calcispheres, which are the only microfos-sils other than foraminifera in the Site 547 washedresidues, have a scattered distribution throughout thesection (Fig. 3).

3.3. Monjas section, SE Spain

Upper Albian and upper Cenomanian sampleswere collected from a section exposed in the Ar-royo del Campillo de las Monjas, North of VelezBlanco in southeastern Spain; locality details are de-scribed by Jansen et al. (1984). A thick sequenceof Aptian=Albian clays and marly clays, which isexposed semi-continuously, is followed by a thin

interval of upper Cenomanian marl–limestone alter-nations, and then by Turonian siliceous limestones.A black shale interval at the Cenomanian=Turonianboundary is absent. The boundaries between thelithologic units correspond to hiatuses (Fig. 4). Theabsence of Ž13C excursions is consistent with thepresence of hiatuses that were recognised from sedi-mentologic and biostratigraphic data.

Microfossils other than foraminifera are ex-tremely rare in our samples, except for one samplewith abundant radiolaria in the Turonian siliceouslimestones (Fig. 4). Comparison of 45 µm and 63µm sieve fractions shows that a large portion ofthe total number of H. moremani is not retained inthe >63 µm sieve fraction, but both fractions givesimilar patterns of stratigraphic variation. The oldestH. moremani are found near the base of the upperAlbian R. ticinensis–subticinensis Zone, and a firstincrease in frequencies occurs within the Rotaliporaappenninica Zone. Above that, the relatively lownumber of samples, and the presence of hiatusesallow only for recognition of moderately high per-centages of H. moremani in Cenomanian samples,and high frequencies of heterohelicids as a group

120A

.J.Nederbragt

etal./Palaeogeography,

Palaeoclimatology,Palaeoecology

144(1998)

115–133Fig. 3. Faunal composition in the >45 µm sieve fraction of samples from DSDP Site 547 with bulk carbonate Ž13C record. Shaded areas give faunal estimates for eachsample, thick lines show a three-point moving average to emphasise general trend. Horizontal lines trace base and top of Ž13C excursion.

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 121

Fig. 4. Percentage of heterohelicids within the planktonic foraminiferal faunas, extent of fragmentation of planktonic foraminifera andbulk sediment Ž13C record in the Monjas section, SE Spain. Microfossils other than planktonic foraminifera are rare; arrow points tosingle sample with abundant radiolaria. Estimates are based on the >63 µm (black dots) and the >45 µm sieve fractions (squares) ofwashed residues, and thin-sections (open circles).

in the Turonian. Guembelitria specimens were notfound in the Monjas samples.

3.4. Cismon section, northern Italy

The Cismon section offers a continuous recordof upper Albian to Turonian limestones and siliceouslimestones deposited in a fully pelagic setting (Chan-nell and Medizza, 1981). The section includes a0.5 m thick black shale, the Bonarelli level, thatrepresents the latest Cenomanian OAE2. Sampleswere taken at 1 m intervals throughout the section(Fig. 5). For this study, the standard R. appenninicaZone is subdivided into a local Planomalina bux-

torfi Zone and a younger R. appenninica Zone. Thetwo species have similar stratigraphic ranges in otherareas (Caron, 1985), but the first occurrence of R.appenninica in the thin-sections from the Cismonsection is found higher in the stratigraphy than ex-pected from its documented first appearance. Bios-tratigraphic results from our samples are in generalagreement with those presented in Bellanca et al.(1996). Minor differences in the position of zonalboundaries can be attributed to the fact that biostrati-graphic results are based on thin-sections, in whichdiagnostic cuts through marker species are relativelyrare, and the statistical chance of finding a zonalmarker is small. The pattern of stratigraphic varia-

122 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

Fig. 5. Faunal composition in thin-sections from the Cismon section in northern Italy. Dark shaded area for heterohelicids and fragmentsgive estimates for all samples with thick lines showing a three-point moving average to emphasise general trend. Note that percentagesof radiolaria, planktonic foraminifera and calcispheres add to 100%. Shaded horizontal bars trace stratigraphic levels of Ž13C excursions.

tion in Ž13C and its correlation with biostratigraphyis in good general agreement with published records(Jenkyns et al., 1994).

Planktonic foraminifera are the most frequent mi-crofossils in the section as a whole, but calcispheresand radiolaria are abundant in various levels. Bel-lanca et al. (1996) found that radiolarian abundancesvaried at the scale of orbital cycles. Abundant cal-cispheres usually occur in levels where severe dis-solution occurred, indicated by abundant fragmentsof planktonic foraminifera (Fig. 5). On the otherhand, the amount of radiolaria and percentage offragments are mostly unrelated, which indicates thathigh abundances of radiolaria are not the result ofenrichment because of dissolution of carbonate. Ra-diolaria disappear from our samples in the lowermostCenomanian at the top of the A=C boundary Ž13Cexcursion. This could be an artefact of the equaldistance sample spacing, if all samples were con-sistently located in the carbonate rich lithologieswithin a cyclic pattern of variation in silica content.

However, sedimentologic description of the section,shown schematically in Fig. 5, indicates that thescarcity of radiolaria in Cenomanian samples is real,as it corresponds to a decrease in the amount of chertin the section.

Heterohelix moremani shows an overall increasein frequencies throughout the section, but with short-term fluctuations (Fig. 5). The percentage of hetero-helicids remains on average more or less constant inthe upper half of the R. appenninica Zone, but is fol-lowed by a minor maximum immediately above theA=C boundary excursion in Ž13C. At the same levelthat radiolarian concentrations drop, numbers per unitsurface area of heterohelicids as well as planktonicforaminifera in general increase abruptly. A more pro-nounced maximum in relative frequencies of H. more-mani occurs in the middle Cenomanian. The variationin abundance of H. moremani does not correlate obvi-ously with the degree of fragmentation in Albian andCenomanian samples. However, fluctuations in rela-tive frequencies of heterohelicids in Turonian lime-

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 123

stones above the Bonarelli level are more likely to bethe result of selective dissolution. Low percentages ofheterohelicids in the Turonian are found in sampleswhere most foraminifera are dissolved.

3.5. Oued Mellegue, Tunisia

A thick sequence of upper Cenomanian andyounger hemipelagic sediments is exposed aroundthe artificial lake in Oued Mellegue, northeasternTunisia, which is in the centre of a faulted anti-cline. The section sampled for this study follows asmall subsidiary gully west of the lake. The OAE2-interval is represented by a 36 m thick sequence oflimestones and marly limestones that contain up to3% organic carbon, the Bahloul Formation (Burol-let, 1956). The ¾2.5‰ Ž13C excursion starts at thebase of the carbonaceous interval (Fig. 6). Estimatesbased on Milankovitch cyclicity for the duration ofthe R. cushmani Zone (2.4 Ma; Gale, 1989), thebase of which is not exposed in the section, and theWhiteinella archaeocretacea Zone (0.7 Ma; Kuhntet al., 1997) give maximum accumulation rates of20 cm=ka for the section as a whole. Other timescales would give lower values. Correlation of thecarbonaceous interval in the Mellegue section withthe Moroccan Tarfaya basin (Kuhnt et al., 1990,1997) indicate that the carbonaceous Bahloul Forma-tion represents accumulation rates of approximately10 cm=ka, which rates are up to twice as low as inthe section as a whole.

The large number of samples from marls as wellas limestones allow for a comparison of faunalcounts in washed residues (>63 µm fraction) andin thin-section (Fig. 6). The two preparation methodsgive similar relative frequencies in Cenomanian sed-iments. Above that, thin-sections give persistentlylower estimates of heterohelicid percentages thanwashed residues for Turonian samples. Presumably,the change from Cenomanian faunas to lower Tur-onian faunas with abundant H. globulosa and large,globular Whiteinella spp. affect estimates of the per-centage of biserial specimens in thin-section in asystematic way.

Heterohelicids reach their maximum abundancewithin the Bahloul Formation, but relative frequen-cies vary with lithology within that interval. Het-erohelicids are relatively rare in marly, bioturbated

levels, which contain abundant fragments (Fig. 6),indicating that dissolution of carbonate occurred. Incontrast, planktonic foraminiferal faunas in lami-nated limestones in the middle of the interval aremuch better preserved, with less severe dissolu-tion indicated by low amount of fragments. Theselaminated beds contain the highest concentration ofheterohelicids in the section, with spiral planktonicforaminifera forming less than 30% of the planktonicforaminiferal assemblages generally.

As far as can be seen from thin-sections, Hed-bergella spp. are the only other species accompa-nying heterohelicids in the black shale samples. Yethigh percentages of heterohelicids in the OAE2 in-terval cannot be explained as relative enrichment,resulting from the disappearance of almost all otherplanktonic foraminiferal species. Concentrations offoraminifera in thin-sections can be converted to arelative measure of rain rates, by multiplication withaccumulation rates. High relative frequencies of het-erohelicids in laminated layers (Fig. 6) correspondto an increase in absolute numbers per unit time,even if accumulation rates for the Bahloul Formationwere twice as low as in the rest of the section. Lowerfluxes in bioturbated levels with abundant fragmentscould be primary, but dissolution of a large portionof the planktonic foraminifera reaching the seaflooras the main cause cannot be excluded.

3.6. Compilation of stratigraphic trends

To compare stratigraphic trends, percentages ofheterohelicids in the four well-dated sections areplotted against numeric time (Fig. 7). No correc-tion is made for sample preparation methods, whichmeans that actual values cannot be compared di-rectly. Comparison of the 45 µm and 63 µm sievesused for the Monjas samples and comparison of ourresults for Site 547 with those of Leckie (1987)show that samples washed over a 45 µm sieve givea higher percentage of H. moremani. Cenomanianthin-sections from the Mellegue section give similarvalues as >63 µm washed residues, i.e. are under-estimating the real proportion of biserial specimens.In theory, even the smallest foraminifera should bevisible in thin-section, but morphology determineswhat proportion of specimens is recognisable. Spiralforms can be recognised as such from any cut in thin-

124A

.J.Nederbragt

etal./Palaeogeography,

Palaeoclimatology,Palaeoecology

144(1998)

115–133

Fig. 6. Faunal composition in thin-sections (open circles) and >63 µm washed residues (black dots) from the Mellegue section, Tunisia.

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 125

Fig. 7. Combined heterohelicid abundance records plotted againstnumeric time scale of Haq et al. (1987).

section approximately through the centre of the test.Biserial forms can only be identified positively fromcuts that are more or less parallel to the long axisof the test. Differences between estimates from thin-sections and washed residues of Turonian samplesfrom the Mellegue section indicate that changes inspecies composition of the planktonic foraminiferalfauna as a whole also affect results.

However, all sections give similar overall trends,and even with different preparation methods, short-

term variations can be recognised that correlate indifferent sections. Heterohelix moremani is moreabundant in upper Albian thin-sections from theCismon section than in the fine sieve fraction ofDSDP Site 457 samples. Yet both sections showan increase in the earliest Cenomanian followingthe A=C boundary Ž13C excursion. The Cismon,Mellegue and Monjas sections give similar percent-ages of H. moremani in thin-sections from the upperCenomanian. The latest Cenomanian OAE2 is notpresent in the Monjas section, and consists of blackshales without foraminifera in the Cismon section(the Bonarelli level). Only the Mellegue section wasdeposited at a shallow enough palaeo-water-depth togive a record of abundance variation in H. moremaniacross the anoxic event. Dominance of heterohelicidsand hedbergellids during this event in the Melleguesection is similar to what was recognised in otherareas (Beckmann, 1972; Kuhnt et al., 1990; Lipson-Benitah et al., 1990). Turonian samples from oxicsediments deposited after the OAE contain abun-dant heterohelicids as a group in all sections. Es-timates for the thin-sections from the Monjas andCismon localities are on average similar to valuesfound in Mellegue thin-sections (Fig. 7), but showstratigraphic variation that correlates with the degreeof fragmentation (Figs. 4 and 5). Washed residuesfrom the Mellegue section show that H. globulosaforms an important part of the Turonian hetero-helicid population, and this species is dissolutionsensitive (Malmgren, 1987). Most likely, variation inabundance in the Turonian limestones of the Cismonand Monjas section should be attributed primarily tovariation in the depth of the lysocline.

3.7. Planktonic foraminiferal oxygen and carbonisotopes

Stable isotope measurements were done on H.moremani and other species in two samples each fromDSDP Sites 137 and 551. Samples from DSDP Site137 are from the early Cenomanian R. brotzeni Zone,while sediments included from Site 551 belong tothe upper part of the R. cushmani Zone (late Ceno-manian). Foraminifera are air-filled, but foraminiferaltests show variable extent of recrystallisation, withaddition of some secondary calcite (Plates I and II).Isotope results are shown in Table 1 and Fig. 8. Het-

126 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

PLATE I

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 127

erohelix moremani is close to Hedbergella delrioensisand to bulk sediment in Ž18O values in all four sam-ples, but the order relative to keeled species varies.The Ž18O contrast amongst planktonic forms is weakat Site 137, but H. moremani is the lightest (warmest)in oxygen isotopes of all planktonic foraminiferalspecies in both samples. It has intermediate Ž18O val-ues in the Site 551 samples, which show a wider rangebetween planktonic foraminiferal species. The pat-tern in Ž13C values is more consistent. Irrespectiveof its Ž18O signature, H. moremani is lighter in Ž13Cthan all other planktonic foraminifera except Guem-belitria cenomana, and close to benthic foraminifera.Light Ž13C values were also found for Cenozoic bise-rial planktonic foraminifera and Palaeogene triserials(Resig and Kroopnick, 1983; Boersma and PremoliSilva, 1989). For further comparison, the Ž13C valueof the modern triserial, Gallitellia vivans, in a boxcore top sample is shown relative to other planktonicforaminiferal species (Table 2). Like mid-CretaceousGuembelitria, G. vivans is the most depleted in Ž13Cof all planktonic foraminifera, even though its Ž18Osignature shows that this species built its test not farbelow the sea surface.

4. Discussion and conclusions

4.1. Temporal abundance trends

All studied sections show an overall increase inabundance of H. moremani from its first appearancein the lower R. ticinensis=subticinensis Zone up to

PLATE I

Black bars are 20 µm; white bars with transmitted light photos are 100 µm.1–6. Heterohelix moremani (Cushman).1. Sample DSDP 137-11-4, 104–108 cm, lower Cenomanian. Note extensive recrystallisation, with addition of calcite indicated by

weakly depressed sutures.2. Specimen with reniform chambers, sample DSDP 547A-41-1, 58–61 cm, lower Cenomanian.3. Specimen with globular chambers, Mellegue section sample 96Me20, upper Cenomanian.4. Edge view of ideotypic specimen, Mellegue section sample 96Me5, middle or upper Cenomanian.5. Side view of ideotypic specimen, Mellegue section sample 96Me5.6. Specimen transitional to H. globulosa in having faint costae, Mellegue section sample 96Me20, upper Cenomanian.7. Transmitted light photo of sample with 25% heterohelicids (two clear specimens indicated by arrows), Cismon section sample

94Cis106, lower Cenomanian.8. Sample from CTBE interval in Mellegue section with dominant heterohelicids, sample 96Me74.9. Thin-section with Heterohelix globulosa, arrow points to specimen with proloculus and all other chambers visible, Mellegue section

sample 96Me123, lower Turonian.

the Cenomanian=Turonian boundary. Heterohelicidsare generally rare up to the uppermost Albian, thencommon to moderately abundant until the uppermostCenomanian, and very abundant from the base of theuppermost Cenomanian OAE2. During the anoxicevent, Heterohelix started to diversify, and hetero-helicids remained an abundant part of planktonicforaminiferal faunas until the end of the Cretaceous(Nederbragt, 1991). The pattern of variation in themid-Cretaceous confirms that biostratigraphic inter-pretation of heterohelicid abundance in Venezuelansections (Fig. 2) is a good approximation, but therelatively small number of samples does not allowfor further refinement of age-interpretations. Moreprecise correlation with other areas might be possi-ble from high-resolution sampling. Although actualabundance values within a given time interval varybetween sections, the results give some indicationthat not only long-term but also short-term fluctua-tions could be synchronous over a wider area. Forexample, H. moremani is more frequent in the upperAlbian in the Cismon section than in age equiva-lent samples from DSDP Site 547, but trends in thelower Cenomanian appear to be parallel (Fig. 7). Thesections included in this study are all from low tomid-latitudes in the Circum-North Atlantic. Furtherdocumentation would have to show if the trend inabundance of H. moremani was global.

4.2. Stable isotope signature and habitat

Faunal variation that is synchronous over a widerarea implies that it is the result of large-scale changes

128 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

PLATE II

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 129

Table 1Single species stable isotope measurements in mid-Cretaceous samples

137-10-3, 136–138 cm 137-11-4, 104–108 cm 551-6-1, 43–45 cm 551-6-2, 15–18 cm

Ž13C Ž18O Ž13C Ž18O Ž13C Ž18O Ž13C Ž18O

Bulk (powdered sediment) 1.93 �3.52 1.97 �3.03 2.40 �1.91 2.40 �1.32Guembelitria cenomana 0.89 �1.50Hedbergella delrioensis 1.83 �3.41 1.96 �3.06 2.16 �1.79 2.25 �1.43Hx. moremani (45–63 µm) �0.53 �4.08 0.01 �3.30 1.23 �1.92Hx. moremani (63–90 µm) 1.47 �1.89 1.56 �1.58Praeglobotruncana spp. 1.90 �3.65 1.97 �3.13 2.61 �2.24Rotalipora cushmani 1.74 �2.18 2.01 �1.86Rotalipora gandolfi 1.55 �3.51 1.55 �2.66Mixed benthic foraminifera 0.22 �2.00 0.40 �1.56 0.89 �0.89 1.42 �0.75

% Hx. moremani (>45 µm) 9.6 5.3 12.7 17.6

in ocean circulation. Information about the habitat ofH. moremani should give an indication where in thewater column changes must have occurred to be ableto affect the distribution of this species. If the rela-tive order of species in Ž18O values is taken at facevalue, then it should be concluded that the depthat which H. moremani lived was variable. Valuesmeasured here (Table 1) are in a range that wasconsidered primary and representative of sea surfacewater temperatures by Sellwood et al. (1994), butthat was interpreted as the result of diagenetic alter-ation by Killingly (1983). Here, no attempt is madeto solve if absolute Ž18O values are primary or not.Diagenesis can cause a systematic shift especially inoxygen isotope values, but in general the relative or-der between species remains preserved (Killingly,1983; Schragg et al., 1992). However, the small,thin-walled test of H. moremani has a large sur-face-to-volume ratio, and precipitation on the wall

PLATE II

Bar is 5 µm.1. Detail of external wall structure of H. moremani specimen in Plate I, fig. 1, Sample DSDP 137-11-4, 104–108 cm, lower Cenomanian.2. Detail of H. delrioensis specimen from which ultimate chamber was removed, showing moderately well preserved wall but with minor

secondary calcite on outer and inner wall surfaces, Sample DSDP 137-11-4, 104–108 cm, lower Cenomanian.3. Detail of moderately well preserved wall structure of R. cushmani specimen, Sample DSDP 137-11-4, 104–108 cm, lower

Cenomanian.4. Broken H. moremani specimen, showing extensive recrystallisation and addition of secondary calcite. Note large euhedral calcite

crystal on the right. Sample DSDP 551-6-1, 43–35 cm, upper Cenomanian.5. Broken H. delrioensis specimen showing recrystallisation, Sample DSDP 551-6-1, 43–35 cm, upper Cenomanian.6. Partly recrystallised wall structure of broken R. cushmani specimen, with minor growth of crystals on inner surface, Sample DSDP

551-6-1, 43–35 cm, upper Cenomanian.

surface of secondary calcite (Plate II) would have alarger impact on the measured stable isotope com-position than in other species with thicker shells.Cenomanian sediments at DSDP Sites 137 and 551are amongst the best preserved available (Sellwoodet al., 1994), but preservation is by no means perfect.Cenomanian sediments at Site 551 were depositedon top of oceanic basement, and interaction withbasement fluids could have caused size dependentenrichment in Ž18O. On the other hand, interspeciescontrast in stable oxygen isotopes at Site 137 aretoo small to allow for conclusive depth-ranking ofplanktonic foraminiferal species, even if values arenot altered. However, a surface water habitat for H.moremani in at least part of its range is supportedby its abundance in shallow water deposits (Leckie,1987). The light Ž13C values found for biserial andtriserial forms are more conclusive. The favouredexplanation here is that H. moremani was most likely

130 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

Fig. 8. Cross-plots of 13C and 18O values of foraminiferal species in mid-Cretaceous DSDP samples and a box core top sample from theBanda Sea; values are listed in Tables 1 and 2.

a surface dweller, maybe lived in deeper waters insome parts of its geographic range, but that it wasdepleted in Ž13C independent of its position in thewater column.

Depletion in 13C relative to other planktonic fora-minifera is found not only for H. moremani, butalso for mid-Cretaceous and Palaeogene Guembeli-tria spp. and Cenozoic biserial planktonic foramini-fera (Fig. 8; Resig and Kroopnick, 1983; Boersmaand Premoli Silva, 1989) and modern Gallitellia vi-vans and Globigerina bulloides (Table 2). Cenozoicbiserial planktonic foraminifera were heavy in stableoxygen isotopes in most of their stratigraphic range,

which led to the hypothesis that these forms buildtheir test in equilibrium with ambient water deep inthe water column in an Oxygen Minimum Zone de-pleted in 13C (Resig and Kroopnick, 1983; Boersmaand Premoli Silva, 1989). However, Ž18O values inthe two living species from the Banda Sea, Gl. vivansand G. bulloides, indicate that there is no correlationbetween their Ž13C value and that of ambient water(Table 2). Both species are depleted in 13C relative toother planktonic foraminifera, for which Ž18O valuesshow that they lived at a similar depth in the watercolumn. More extensive data for G. bulloides indi-cate that this opportunistic species incorporates car-

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 131

Table 2Stable isotope measurements for modern foraminifera fromBanda Sea box core (after Kroon and Nederbragt, 1990)

G5-2-33B Ž13C Ž18O

Globigerina bulloides �1.26 �2.70Globigerinoides trilobus 1.60 �2.44Gallitellia vivans �2.35 �2.16Globorotalia menardii 1.16 �1.38Uvigerina perigrina �0.84 3.11Globobulimina pacifica 1.48 3.84

bon out of equilibrium with the CO2 pool throughoutits life span (Spero, 1991; Kroon and Darling, 1995).Other planktonic foraminiferal species also have alight Ž13C signal initially, which was attributed tohigher metabolic rates in juveniles (Berger and Vin-cent, 1986; Kroon and Darling, 1995). A light Ž13Csignature as a result of rapid growth in G. bulloidesthroughout its life span is consistent with the factthat this species is an opportunist that proliferates inhigh-productivity areas. Mid-Cretaceous biserial andtriserial planktonic foraminifera are similar to G. bul-loides, in that the their Ž13C values are consistentlylighter than other planktonic groups with nearly thesame Ž18O value. Most likely, their light Ž13C sig-nature reflects ‘vital effects’ as those in G. bulloidesrather than chemical composition of the ambient wa-ter. It is tempting to equate such a light Ž13C signaturein fossil species with an opportunistic life style.

There can be little doubt that H. moremani wasan opportunist. It had maximum abundance duringthe latest Cenomanian OAE2, when most other spi-ral planktonic foraminifera were greatly reduced innumbers or, in the case of Rotalipora species, hadbecome extinct. Data from the Mellegue section(Fig. 6) show that the increase in percentages withinthe population were not only the result of disappear-ance of other forms. Heterohelix moremani was notmerely surviving, but the increase in fluxes indicatesthat it was actually thriving.

4.3. Palaeoproductivity

Even if an opportunist, H. moremani is not nec-essarily a straightforward palaeo-high-productivityindicator. An obvious tracer for any relation be-tween variation in abundance of this species and

surface water productivity would be to measure theŽ13C-contrast between planktonic and benthic fora-minifera in a deep water section. However, such arecord would be hard to obtain in practice, as few,if any, mid-Cretaceous deep water sections are wellenough preserved to establish a continuous strati-graphic record of 13C measurements on individualspecies. But sedimentologic and faunal data offersome indication that increased abundances of H.moremani during the Cenomanian may correspondto lower productivity overall. At DSDP Site 547(Fig. 3), the overall increase of H. moremani is ac-companied by an overall decrease in Guembelitria,which was classified as a high-productivity indica-tor (Kroon and Nederbragt, 1990). In the Cismonsection, high abundances in the Cenomanian occurafter a distinct decrease in the abundance of radio-laria. In the modern ocean, radiolaria are abundantin areas with high primary productivity. In addition,on a global scale, burial rates of phosphate droppedfrom the late Albian to the end of the Cenoma-nian (Follmi, 1996), which, since phosphate burial islinked to burial of organic matter, may point to anoverall decrease in productivity. None of these argu-ments is conclusive on its own, but taken togetherthey suggest that an increase in abundance of H.moremani cannot be interpreted as a straightforwardindication of increased palaeoproductivity. Heterohe-lix moremani appears to have been a surface dwellingopportunist, that could withstand extreme conditions.Living planktonic foraminifera thrive at intermediatelevels of productivity. Presumably, as an opportunist,H. moremani could withstand higher levels of pro-ductivity than most other planktonic foraminiferalspecies. Yet assemblages with abundant H. moremaniprobably represent only moderately high palaeopro-ductivity. Its abundance variation at DSDP Site 547and in the Mellegue section correlates with excur-sions in the stratigraphic Ž13C record. Such positiveexcursions are attributed to enrichment of the global13C reservoir due to excess burial of organic mate-rial, either through increased preservation of organiccarbon or through enhanced productivity, or to acombination of the two (Pedersen and Calvert, 1990;Arthur and Sageman, 1994). This indicates a rela-tion with global oceanographic changes, that shouldbe taken into account when interpreting variation inabundance of H. moremani on a regional scale.

132 A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133

Acknowledgements

We are grateful to J.E. van Hinte, D. Kroon, I.Premoli Silva and F. Robaszynski for suggestions forimproving earlier versions of the manuscript. Thanksare due to shipboard crew and onshore staff of DSDPand ODP for making part of the samples available.S. Kars made the S.E.M. photographs. This is apublication of the Netherlands Research School ofSedimentary Geology.

References

Alberdi, M., Lafargue, E., 1993. Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous), InteriorMountain Belt, Eastern Venezuela. Org. Geochem. 20, 425–436.

Arthur, M.A., Sageman, B.B., 1994. Marine black shales: de-positional mechanisms and environments of ancient deposits.Annu. Rev. Earth Planet. Sci. 22, 499–551.

Barron, E.J., 1987. Cretaceous plate tectonic reconstructions.Palaeogeogr., Palaeoclimatol., Palaeoecol. 59, 3–29.

Barron, E.J., Harrison, C.G.A., Sloan, J.L., Hay, W.W., 1981.Paleogeography, 180 million years ago to the present. EclogaeGeol. Helv. 74, 443–470.

Beckmann, J.P., 1972. The foraminifera and some associatedmicrofossils of Site 135 to 144 . Init. Rep. DSDP 14, 389–420.

Bellanca, A., Claps, M., Erba, E., Masetti, D., Neri, R.,Premoli Silva, I., Venezia, F., 1996. Orbitally inducedlimestone=marlstone rhythms in the Albian–Cenomanian Cis-mon section (Venetian region, northern Italy): sedimentol-ogy, calcareous and siliceous plankton distribution, elemen-tal and isotope geochemistry. Palaeogeogr., Palaeoclimatol.,Palaeoecol. 126, 227–260.

Berger, W.H., Vincent, E., 1986. Deep-sea carbonates: readingthe carbon-isotope signal. Geol. Rundsch. 75, 249–269.

Boersma, A., Premoli Silva, I., 1989. Atlantic Paleogene biserialheterohelicid foraminifera and oxygen minima. Paleoceanog-raphy 4, 271–286.

Burollet, P.F., 1956. Contribution a l’etude stratigraphique de laTunisie Centrale. Annales des Mines et de la Geologie, Tunis18, 345 pp.

Caron, M., 1985. Cretaceous planktic foraminifera. In: Bolli,H.M., Saunders, J.B., Perch-Nielsen, K. (Eds.), PlanktonStratigraphy. Cambridge University Press, Cambridge, pp. 17–86.

Channell, J.E.T., Medizza, F., 1981. Upper Cretaceous andPalaeogene magnetic stratigraphy and biostratigraphy fromthe Venetian (Southern) Alps. Earth Planet. Sci. Lett. 55, 419–432.

Direccion de Geologıa, 1970. Lexico estratigrafico de Venezuela.Bol. Geol. Publ. Esp. 4, 756 pp.

Erbacher, J., Thurow, J., Littke, R., 1996. Evolution patterns of

radiolaria and organic matter variations; a new approach toidentify sea-level changes in mid-Cretaceous pelagic environ-ments. Geology 24, 499–502.

Follmi, K., 1996. 160 m.y. record of marine sedimentary phos-phorus burial: Coupling of climate and continental weatheringunder greenhouse and icehouse conditions. Geology 23, 859–862.

Gale, A.S., 1989. A Milankovitch scale for Cenomanian time.Terra Nova 1, 420–425.

Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctu-ating sea levels since the Triassic. Science 235, 1156–1167.

Hedberg, H.D., Pyre, A., 1944. Stratigraphy of northeasternAnzoategui, Venezuela. Am. Assoc. Pet. Geol. Bull. 28, 1–28.

Hinz, K., Winterer, E.L., et al., 1984. Initial Reports of the DeepSea Drilling Project, 79. U.S. Government Printing Office,Washington, DC, 934 pp.

Jansen, H.S.M., Kroon, D., van Hinte, J.E., 1984. Entry and exitof Planomalina buxtorfi in a section North of Velez Blanco,southeastern Spain. Rev. Esp. Micropaleontol. 16, 381–397.

Jenkyns, H.C., Gale, A.S., Corfield, R.M., 1994. Carbon- andoxygen-isotope stratigraphy of the English Chalk and ItalianScaglia and its paleoclimatic significance. Geol. Mag. 131,1–34.

Killingly, J.S., 1983. Effects of diagenetic recrystallization on18O=16O values of deep-sea sediments. Nature 301, 594–597.

Koutsoukos, E.A.M., 1994. Early stratigraphic record and phy-logeny of the planktonic genus Guembelitria Cushman, 1933.J. Foram. Res. 24, 288–295.

Kroon, D., Darling, K., 1995. Size and upwelling control ofthe stable isotope composition of Neogloboquadrina dutertrei(d’Orbigny) and Globigerina bulloides d’Orbigny: examplesfrom the Panama Basin and Arabian Sea. J. Foram. Res. 25,39–52.

Kroon, D., Nederbragt, A.J., 1990. Ecology and paleoecology oftriserial planktic foraminifera. Mar. Micropaleontol. 16, 25–38.

Kuhnt, W., Herbin, J.P., Thurow, J., Wiedmann, J., 1990. Distri-bution of Cenomanian–Turonian organic facies in the WesternMediterranean and along the adjacent Atlantic margin. In:Huc, A.Y. (Ed.), Deposition of Organic Facies. Am. Assoc.Pet. Geol. Stud. Geol. 30, 133–160.

Kuhnt, W., Nederbragt, A.J., Leine, L., 1997. Cyclicity ofCenomanian–Turonian organic-carbon-rich sedimentation inthe Tarfaya coastal basin (Morocco). Cretaceous Res. 18,587–601.

Leckie, R.M., 1984. Mid-Cretaceous Planktonic foraminiferalbiostratigraphy off central Morocco, Deep Sea Drilling ProjectLeg 79, Sites 545 and 547 . Init. Rep. DSDP 79, 579–620.

Leckie, R.M., 1987. Paleoecology of mid-Cretaceous planktonicforaminifera: a comparison of open ocean and epicontinentalsea assemblages. Micropaleontology 33, 164–176.

Li, Q., Radford, S.S., 1992. Morphology and affinity of theplanktonic foraminifer Cassigerinelloita amekiensis Stolk andreclassification of Cassigerinelloita Stolk. Proc. ODP, Sci.Results 120, 595–600.

Lipson-Benitah, S., Flexer, A., Rosenfeld, A., Honigstein, A.,Conway, B., Eris, H., 1990. Dysoxic sedimentation in the

A.J. Nederbragt et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 144 (1998) 115–133 133

Cenomanian–Turonian Daliya Formation, Israel. In: Huc, A.Y.(Ed.), Deposition of Organic Facies. Am. Assoc. Pet. Geol.Stud. Geol. 30, 27–39.

Malmgren, B.A., 1987. Differential dissolution of Upper Creta-ceous planktonic foraminifera from a temperate region of theSouth Atlantic Ocean. Mar. Micropaleontol. 11, 251–272.

Nederbragt, A.J., 1991. Late Cretaceous biostratigraphy anddevelopment of Heterohelicidae (planktic foraminifera). Mi-cropaleontology 37, 329–372.

Pedersen, T.F., Calvert, S.E., 1990. Anoxia vs. productivity: whatcontrols the formation of organic-carbon-rich sediments andsedimentary rocks? Am. Assoc. Pet. Geol. Bull. 74, 454–466.

Resig, J.M., Kroopnick, P.M., 1983. Isotopic and distributionalevidence of a planktonic habit for the foraminiferal genusStreptochilus Bronnimann and Resig, 1971. Mar. Micropale-ontol. 8, 235–248.

Scholle, P.A., Arthur, W.A., 1980. Carbon isotope fluctuationsin Cretaceous pelagic limestones: potential stratigraphic andpetroleum exploration tool. Am. Assoc. Pet. Geol. Bull. 64,67–87.

Schragg, D.P., DePaolo, D.J., Richter, F.M., 1992. Oxygen iso-tope exchange in a two-layer model of oceanic crust. EarthPlanet. Sci. Lett. 111, 305–317.

Sellwood, B.W., Price, G.D., Valdes, P.J., 1994. Cooler estimatesof Cretaceous temperatures. Nature 370, 453–455.

Spero, H.J., 1991. Experimental evidence for stable isotopic dise-quilibria in the planktonic foraminifera Globigerina bulloides.EOS 72, 155–156.

Yoris, V.F., 1990. Localidades tipo para los miembros de laFormacion Chimana en la Serranıa del Interior VenezuelaNororiental. GEOS 30, 295–324.