The Platycopid Signal deciphered: Responses of ostracod taxa to environmental change during the Cenomanian–Turonian Boundary Event (Late Cretaceous) in SE England

Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p a l a e o

The Platycopid Signal deciphered: Responses of ostracod taxa to environmental change during the Cenomanian–Turonian Boundary Event (Late Cretaceous) in SE England David J. Horne a, b,⁎, Simone Nunes Brandão c, d, Ian J. Slipper e a

Queen Mary University of London, School of Geography, Mile End Road, London E1 4NS, UK The Natural History Museum, Dept of Zoology, Cromwell Road, London SW7 5BD, UK c Biozentrum Grindel und Zoologisches Museum, Universität Hamburg, Martin-Luther-King Platz 3, 20146 Hamburg, Germany d Deutsches Zentrum für Marine Biodiversitätsforschung, Senckenberg Forschungsintitut, Südstrand 44, 26382 Wilhelmshaven, Germany e School of Science, University of Greenwich at Medway, Chatham Maritime, Kent, ME4 4TB, UK b

a r t i c l e

i n f o

Article history: Received 17 February 2011 Received in revised form 17 May 2011 Accepted 19 May 2011 Available online 27 May 2011 Keywords: Platycopid Signal Ostracoda Oceanic Anoxic Events Cretaceous Oligotrophy

a b s t r a c t A multi-proxy investigation of the Cenomanian–Turonian Boundary Event (CTBE) near Dover in southern England previously demonstrated that, coincident with a major global positive carbon stable-isotope excursion during Oceanic Anoxic Event 2 (OAE 2), the diversity of nanoplankton, dinoflagellates, planktonic and benthonic foraminifera and ostracods was severely reduced. This was attributed to decreasing levels of dissolved oxygen, consequent on an intensification and expansion of the oceanic Oxygen Minimum Zone (OMZ) into shelf seas. In the case of the ostracods, it was noted that as podocopid taxa became locally extinct, platycopids became the dominant component of the fauna. Subsequently the “Platycopid Signal” Hypothesis (PSH) claimed that dominance of platycopids in ostracod assemblages could be regarded as a signal of dysaerobic conditions on the sea floor, based on the premise that the filter-feeding platycopids, able to pass more water over their respiratory surface, were better-equipped than other benthonic ostracods to survive in water of reduced oxygen concentration. The PSH has been widely accepted and applied in palaeoenvironmental reconstructions of stratigraphic intervals ranging from the Palaeozoic to the Quaternary. However, the modern biological and ecological support claimed for the Platycopid Signal has been challenged; platycopids are occasionally dominant in modern OMZs, but often they are not, and in any case the same can be said about some podocopids. Apparently precise calibration scales published by some authors are not justified by available data; furthermore, Platycopid Signal indications of dysaerobic intervals in the English Chalk succession often conflict with the evidence of macrofossils and trace fossils. Here we review old and new data from two CTBE sites in SE England, Dover and Eastbourne, and advance an alternative interpretation of the Platycopid Signal, based on the concept of the spread of oceanic oligotrophic conditions into the European Chalk Sea during OAE2. We propose that ostracod assemblages overwhelmingly dominated by platycopids signify oligotrophy, because living platycopids appear to be adapted to filter-feed on nano- and picoplankton which are predominant in oligotrophic conditions. © 2011 Elsevier B.V. All rights reserved.

1. Introduction According to the Platycopid Signal Hypothesis (PSH) of Whatley (1991), ostracod assemblages dominated by species belonging to the Order Platycopida can be interpreted as indicating dysaerobic conditions on the sea floor, thus offering a valuable tool for the study of Oceanic Anoxic Events (OAEs). Platycopids, as filter feeders, were considered by Whatley to be more efficient at circulating water through the domiciliar space inside the carapace, for both feeding and ⁎ Corresponding author at. Queen Mary University of London, School of Geography, Mile End Road, London E1 4NS, UK. E-mail addresses: [email protected] (D.J. Horne), [email protected], [email protected] (S.N. Brandão), [email protected] (I.J. Slipper). 0031-0182/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2011.05.034

respiration purposes, and thus better-adapted than non-filter feeders to survive at low levels of dissolved oxygen (Whatley, 1991; Brandão and Horne, 2009). Originally developed on the basis of data from the Cenomanian–Turonian Boundary Event (CTBE) (also known as OAE 2) in SE England, the PSH has been applied to the interpretation of ostracod assemblages from the Quaternary to the Palaeozoic (Boomer and Whatley, 1992; Whatley and Arias, 1993; Whatley et al., 1994; Aiello et al., 1996; Steineck and Thomas, 1996; Majoran et al., 1997; Majoran and Widmark, 1998; Boomer, 1999; Gebhardt, 1999a; Crasquin-Soleau and Kershaw, 2005; Bergue et al., 2007). The PSH was developed beyond its early concept by Lethiers and Whatley (1994, 1995) who applied it to Late Palaeozoic ostracod assemblages in NW Europe and published comparative tables (e.g., Lethiers and Whatley, 1994, Fig. 2) in which the percentages of

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

platycopid species in assemblages were matched to ranges of dissolved oxygen levels. Their analyses showed a good match (on a broad scale at least) with other proxy evidence indicative of, for example, low oxygen events in the Late Devonian and welloxygenated, cooler waters associated with Early Carboniferous glaciation. However, it must be noted that while Lethiers and Whatley followed Adamczak (1969) in assuming all palaeocope ostracods to have been filter-feeders, Olempska (2008) presented and interpreted new morphological evidence (in particular adductor muscle, frontal and mandibular scars) from Silurian, Devonian and Carboniferous taxa and concluded that beyrichioidean palaeocopes (including paraparchitids) were not filter-feeders, but actively-feeding scavengers or carnivores. It remains to be seen what impact this reinterpretation of ostracod feeding modes may have on the PSH as applied to Palaeozoic assemblages. Whatley (1995) presented a new version of his calibration table, with percentages of filter-feeders based on the relative abundance of individuals, not species as in the earlier version published by Lethiers and Whatley (1994, 1995). Subsequently these comparative values were refined and presented as a scale with an apparent precision of 0.5 ml/l dissolved oxygen (Whatley et al., 2003), reproduced herein as Table 1. In none of the above-cited papers was a full explanation given of how the calibrations shown in the tables were achieved; for example, Whatley et al. (2003: p. 360) simply stated that their improved table was based “…mainly on further considerations of the available literature on Recent ostracod ecology.” In the same paper they went on to apply the PSH to the estimation of dissolved oxygen levels in the Late Cretaceous, giving values with a higher precision (0.05 ml/l) than that of their calibration table, which suggests the use of a regression line fitted to a data plot. While this technique might be implied it was not made explicit and consequently their method constitutes a “black box” that is not open to scrutiny by readers. This in itself does not mean that the method is not valid, but it makes it very difficult to test its validity. Almost any fossil proxy is capable of yielding what appear to be plausible results; it is important to compare such results with those of other proxies to see whether they agree or disagree. In some intervals at least, the results of Whatley et al.'s (2003) application of the PSH to the Late Cretaceous Chalk of Norfolk stand in isolation from those provided by macrofossils and trace fossils. To take just one example, the PSH led them to identify “a notable period of dysaerobia in the later part of the Weybourne Chalk (including the Catton Sponge Beds) culminating in an oxygen trough in the Beeston Chalk.” (Whatley et al., 2003: 365). This is difficult to sustain, given the total fauna recorded through this interval; according to Mortimore et al. (2001) the Catton Sponge Bed (Hardground II) contains a rich assemblage of hexactinellid sponges together with moulds of originally aragoniteshelled bivalves and gastropods, and is penetrated by an extensive Thalassinoides burrow system. Inoceramid bivalves, echinoids, foraminifera, belemnites and the ammonites have all been recovered from the associated levels. This is evidence of a well-oxygenated environ-

Table 1 Scale of actual oxygen values related to the percentage of species of filter-feeding ostracods in an assemblage; after Whatley et al. (2003: p. 360), with correction of an error on the original which showed 20–30% platycopids in the second cell from the bottom in the right-hand column. >90% platycopids = <1.5 ml/l 80–90% platycopids = 2–1.5 ml/l 70–80% platycopids = 2.5–2 ml/l 60–70% platycopids = 3–2.5 ml/l 50–60% platycopids = 3.5–3 ml/l 40–50% platycopids = 4–3.5 ml/l 30–40% platycopids = 4.5–4 ml/l 20–30% platycopids = 5–4.5 ml/l <20% platycopids = >5 ml/l

80–>90% platycopids = very low oxygen 2–<1 ml/l

305

ment above, at and below the sediment water interface. The Beeston Chalk, which according to Whatley et al. (2003) contains an oxygen trough, has yielded a high-diversity macrofauna and is rich in microfossils (Mortimore et al., 2001)—not the kind of assemblage normally associated with low oxygen levels. In contradiction of the PSH, Gebhardt (1999b) found the platycopid Cytherella associated with podocopids in well-oxygenated palaeoenvironments in the Cenomanian–Turonian marine shelf sediments of NE Nigeria, while species of the podocopids Cythereis and Ovocytheridea characterised a group preferring normal oxygen concentrations (associated with limestones) but somewhat tolerant of lower levels (indicated by laminated black shales). Gebhardt and Zorn (2008), in a study of Cenomanian ostracod assemblages from the Tarfaya region of Morocco, noted that two genera were better adapted than others to oxygen depletion: the podocopid Reticulocosta and the platycopid Cytherelloidea, the former being the more abundant and the latter being absent from intervals of severe oxygen deficiency and/ or deeper water. Commenting that their findings did not fully agree with Whatley's PSH, as podocopids always dominated their assemblages, they concluded that Cytherelloidea was not an indicator of very low oxygen levels. There are two main ways in which the PSH can be tested: 1. Its uniformitarian basis can be tested by studying the biology, ecology and distribution of living marine ostracods to determine how well they support the claim that platycopids are dominant in OMZs. 2. Its effectiveness in application to fossil assemblages can be tested by comparing its predictions with the indications of other proxies for past oxygen levels, such as other microfossils, macrofossils and trace fossils. The uniformitarian basis of the PSH has recently been questioned (Smith and Horne, 2002; Boomer et al., 2003; Horne, 2003, 2005); a detailed critique has concluded that there is no convincing support for it in modern distributional data and that the apparent precision of the calibration tables is unjustified (Brandão and Horne, 2009). However, this does not necessarily mean that the basic premise of the PSH is wrong; as Brandão and Horne (2009) observed in their conclusions, living marine ostracods are descended from taxa which survived OAEs in the past, so the responses of modern assemblages to changing oxygen levels may not be a reliable guide to how they responded in the Cretaceous. Here we examine and test the effectiveness of the PSH in determining past oxygen levels by comparing its indications with those from other proxies, using data from the CTBE in SE England. 2. Material and methods We have used datasets from two CTBE localities in SE England (Fig. 1). One, from Abbots Cliff, near Dover (National Grid Reference TR 268385; Latitude 51° 06″ 04′ North; Longitude 1° 14″ 19′ East), comprises the ostracod assemblages described by Jarvis et al. (1988). The other comprises new ostracod data from the more expanded equivalent section at Eastbourne; samples were collected at the Beachy Head (Gun Gardens) section (National Grid Reference TV 588953; Latitude 50° 44″ 12′ North; Longitude 0 o 15″ 01′ East) and processed by the same method as those of Jarvis et al. (1988) from Dover, using hydrogen peroxide followed by wet-sieving and picking ostracods from the N63 b 3000 μm fraction. 3. Ostracod assemblages in the CTBE at Dover

60–80% platycopids = low oxygen 3–2 ml/l 40–60% platycopids = medium oxygen 4–3 ml/l 20–40% platycopids = high oxygen 5–4 ml/l <20% platycopids = very high oxygen above 5 ml/l

In their detailed study of the Cenomanian–Turonian boundary interval near Dover, Jarvis et al. (1988) were the first to demonstrate the distinctive sequence of changes in the ostracod fauna in which Whatley (1990, 1991) initially recognized the Platycopid Signal. Jarvis et al. (1988) illustrated the occurrence of ostracod species through the

306

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

Fig. 1. Location map showing palaeogeography of the Anglo-Paris Basin in (A) the Mid Cenomanian and (B) the Early Turonian, and showing three localities where a Platycopid Signal has been recorded in the Cenomanian–Turonian Boundary Event. Modified from Gale et al. (2000).

Plenus Marl Formation (subdivided into beds 1–8) and the immediately overlying “Melbourn Rock Beds” in the section exposed at Abbots Cliff, about 5 km West of Dover (not the Shakespeare Cliff section, as incorrectly stated by Whatley (1995) and Whatley et al. (2003), which is closer to Dover). They demonstrated the progressive disappearance of podocopids and the contrasting persistence of platycopids; five platycopid species range right through the Plenus Marl, while podocopid diversity falls from 10 species in Bed 1 to just one in Bed 8. However, this apparently simple pattern is complicated, after the initial loss of podocopids (four Bed 1 species do not make it into Bed 2), by a temporary reversal of the trend (increasing to eight and seven species in beds 3 and 4 respectively). Nevertheless the

overall picture is clear: through the OAE2 interval represented by the Plenus Marl the diversity of podocopids was reduced to a single species, while platycopids were apparently able to thrive: this is the Platycopid Signal. The single surviving podocopid species in the platycopid-dominated Bed 8 assemblage is the cytheroidean Isocythereis elongata (not, as stated incorrectly by Whatley (1991, 1995) and Whatley et al. (1994, 2003), Imhotepia euglyphea, a species recorded by Jarvis et al. (1988) only in Bed 1). The post-OAE recovery of the ostracod fauna at Dover was documented by Horne et al. (1990) and Slipper (1996); the latter showed the return to pre-OAE diversities to be more rapid than was originally thought, occurring in the earliest Turonian, approximately equivalent with the start of

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

307

Table 2 Numbers of ostracod specimens recovered from samples across the Cenomanian–Turonian boundary at Dover (Abbots Cliff to Akers Steps), SE England (original data used by Jarvis et al., 1988). Samples yielding fewer than 40 specimens have been excluded. Sample

Platycopida

Podocopida

TOTAL

Bairdoppilata spp.

Pontocyprella spp.

Cytheroidea

Others

AKS3 AKS2 AKS1 ABC19 ABC14 ABC13 ABC12 ABC11 ABC10 ABC9 ABC8 ABC7 ABC4

42 41 57 135 61 140 169 112 75 21 250 230 34

2 2 1 6 3 24 19 32 41 85 88 84 12

44 43 58 141 64 164 188 144 116 106 338 314 46

1 0 0 1 0 0 0 7 11 26 2 2 1

0 0 0 0 0 0 0 8 12 10 59 43 9

1 2 1 5 3 24 19 17 18 49 27 39 2

0 0 0 0 0 0 0 0 0 0 0 0 0

the recovery phase of the δ 13C excursion, although the relative abundance of platycopids remained high (30–60%). Jarvis et al. (1988) illustrated only the presence/absence of ostracod species, although comments on relative abundance were made in the text. Here (Table 2, Fig. 2) we present, for the first time, the percentage composition of ostracod assemblages in the CTBE at Dover, using the original data of Jarvis et al. (1988). It is apparent that platycopids were already relatively abundant in pre-OAE Grey Chalk (N70%) and that the first response of the platycopids to the OAE (as indicated by the first buildup phase of the δ 13C excursion) was not an increase, but a marked decrease in relative abundance, from around 75% in Plenus Marl Bed 1 to less than 25% in Bed 2. The corresponding

increase in podocopid relative abundance is particularly seen in bairdioideans and cytheroideans. It is only higher in the Plenus Marl (Bed 6 and upwards) that the Platycopid Signal is evident, coincident with the second buildup and plateau phases of the δ 13C excursion, with platycopids exceeding 90% of the ostracod assemblage. 4. Ostracod assemblages in the CTBE at Eastbourne Paul et al. (1999: Fig. 13) showed the proportions and numbers per 100 g sample of podocopids and platycopids throughout the CTBE interval at Eastbourne. Citing Whatley's PSH they commented that the proportion of platycopids increased “substantially from a

Fig. 2. The Platycopid Signal in the Cenomanian–Turonian Boundary Event at Dover (Abbots Cliff to Akers Steps), SE England. (A) Stratigraphy: columns, from left to right, show stages, ammonite zones (devon. = devonense), planktonic foraminifera zones and lithostratigraphical formations (with the Plenus Marl subdivided into numbered beds) and sample numbers. (B) Outline lithological log. (C) Carbon stable isotope signal, with characteristic intervals labelled on the right-hand side. (D) Summary ostracod data, as percentages of total assemblage in each sample, showing (left) relative proportions of platycopids and podocopids (total counts of specimens on which percentages are based are indicated on the left-hand side), and (right) relative proportions of three podocopid groups. (A) to (C) are redrawn from Figs. 2 and 26 of Jarvis et al. (1988) with modifications based on Gale (1996), Voigt et al. (2006) and Jarvis et al. (2006). The ostracod data are from the original material of Jarvis et al. (1988).

308

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

Fig. 3. The Platycopid Signal in the Cenomanian–Turonian Boundary Event at Eastbourne (Gun Gardens), SE England. (A) Stratigraphy: columns, from left to right, show stages, ammonite zones, planktonic foraminifera zones and lithostratigraphical formations (with the Plenus Marl subdivided into numbered beds) and sample numbers. (B) Outline lithological log. (C) Carbon stable isotope signal, with characteristic intervals labelled on the right-hand side. (D) Summary ostracod data, as percentages of total assemblage in each sample, showing (left) relative proportions of platycopids and podocopids (counts of specimens on which percentages are based are indicated on the left-hand side), and (right) relative proportions of three podocopid groups. (A) to (C) are redrawn from Fig. 4 of Paul et al. (1999) with modifications from Voigt et al. (2006). The ostracod data are original and previously unpublished; because of lateral variation in thicknesses along the section our original lithological log does not correspond precisely to that of Voigt et al.; accordingly the vertical positions of some of our ostracod samples have been adjusted slightly to match them to the correct beds or positions within beds of the Plenus Marl.

65% platycopids in Meads Marl 3, the proportion of platycopids having actually decreased (in the interval represented by Plenus Marl beds 7 and 8 and the overlying lower two thirds of the Ballard Cliff Member) from a maximum of N80% in Plenus Marl Bed 6; their maximum of N90% occurs much higher up, in Holywell Marl 4. Interestingly, their data show the proportion of bairdioidean podocopids (Bairdoppilata spp.) in the Plenus Marl to be at a minimum (0%) in the middle of Bed 1, although their presence below and above that horizon is fairly substantial, reaching a maximum of N60% at the base of bed 3. In Fig. 3 and Table 3 we present our new ostracod data from the CTBE interval at Eastbourne. There are numerous discrepancies between our

minimum of about 10% in basal bed 3 of the Plenus Marl, to a maximum of over 90% in Meads Marl 3” (Paul et al., 1999: 114), omitting to mention two more platycopid peaks lower down, clearly evident in their Fig. 13: one in the middle of Bed 1 of the Plenus Marl (approx. 70%) and the other in their lowest sample, in the underlying Grey Chalk (approx. 65%). Their Fig. 13 actually shows the percentage of Cytherella spp., not total platycopids, and the caption notes that this corresponds very closely to the Platycopid Signal; additional platycopids in the form of Cytherelloidea spp. appear to make up only very small proportions of assemblages. However, there are discrepancies between their text and what is shown in their Fig. 13, which indicates approximately

Table 3 Numbers of ostracod specimens recovered from samples across the Cenomanian–Turonian boundary at Eastbourne (Gun Gardens), SE England. “Others” are species of Cardobairdia (a sigillioidean podocopid) and Polycope (a myodocopid). Sample

Platycopida

Podocopida

TOTAL

Bairdoppilata spp.

Pontocyprella spp.

Cytheroidea

Others

EGG21 EGG17 EGG16 EGG14 EGG12 EGG11A EGG11 EGG10 EGG9 EGG8 EGG5

343 125 231 291 166 169 191 177 234 181 168

11 7 72 54 166 192 140 150 91 157 159

354 132 303 345 332 361 333 329 325 338 327

0 2 15 0 47 102 29 18 1 2 36

0 1 0 0 4 4 51 55 43 50 7

11 4 57 54 115 86 58 77 44 105 114

0 0 0 0 0 0 4 2 3 0 2

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

data and those of Paul et al. (1999), at least some of which may be attributed to the facts that the thicknesses (and very probably the faunas) of the various units vary along the section and that our samples were not taken at identical positions and horizons. Their higher sampling frequency (18 compared to our nine in the Plenus Marl interval) undoubtedly reveals more detail in the vertical changes of ostracod assemblages. The differences may also reflect the different processing methods used; although precisely how this would affect the different taxa is not clear. Our samples were processed by the same method (hydrogen peroxide) as that used by Jarvis et al. (1988) for the Dover study, which has been shown to be unsatisfactory for the more indurated chalks (since it can introduce bias through the loss of thinner-shelled ostracods by dissolution) but acceptable for marls (Slipper, 1996). Paul et al. (1999) used a freezethaw/glauber salts technique which Slipper (1996), in his comparisons of methods, found to be very effective. Probably most influential, however, is the difference in size fractions from which the ostracods were picked. We picked ostracods from a single N63b 3000 μm fraction of each sample residue, as did Jarvis et al. (1988) at Dover. We found a species of the cytheroidean podocopid Amphicytherura to be relatively abundant in our samples from the Grey Chalk and the base of Bed 1 of the Plenus Marl, (11% of the assemblage in the base of Plenus Marl Bed 1), yet this genus was not recorded at all by Paul et al. (1999). The small size of this species (b390 μm long, 230 μm high) means that it would quite easily pass through a 250 μm sieve; Paul et al.'s procedure of picking foraminifera and ostracods from a coarse fraction of N250 μm and a fine fraction of 63– 90 μm (apparently ignoring the 90–250 μm fraction) is likely to have resulted in a failure to record small specimens, not only of this and several other small podocopid species, but also of small juvenile platycopids which were abundant in the majority of our samples. We suspect that the net result of Paul et al.'s procedure was to under-estimate the proportions of platycopids in assemblages. The importance of picking a finer fraction (b250 μm to N125 μm) to obtain good representations of ostracod faunas has been stressed by Weaver (1981) and Horne and Slipper (1992). The above-mentioned discrepancies notwithstanding, both our new data and those of Paul et al. (1999) from Eastbourne show essentially the same overall pattern of changes in ostracod assemblages as was recorded at Dover: an initially high percentage of platycopids, followed by a low and then an increase culminating in very high percentages. The same features of the ostracod faunal turnover in the Plenus Marl have also been recorded by Johnson (1996) at another English CTBE locality, Compton Bay on the Isle of Wight (Fig. 1). It is clear, therefore, that while the changes in the relative abundance of platycopids during the CTBE are perhaps not as simple as has often been stated, the Platycopid Signal undoubtedly exists; the question is, what does it mean? 5. Interpreting the Platycopid Signal If the view that the CTBE interval at Dover and Eastbourne represents the encroachment of the oceanic OMZ into European shelf seas (e.g., Jarvis et al., 1988) is correct, then the PSH appears to be supported, although an explanation is needed for the interval with low platycopid percentages. According to Whatley et al.'s (2003; p. 360) calibration the signal in the CTBE would mean low to medium oxygen levels in the pre-OAE and initial δ13C buildup interval, high (Dover) or medium (Eastbourne) oxygen levels approximately centred on the trough phase of the δ13C signal, and low to very low oxygen levels coincident with the second δ13C buildup and plateau phases. However, as more evidence has been obtained, the productivity-related expanding OMZ model for the OAE has been increasingly questioned and modified by various authors. Some authors argue for increasing productivity (resulting in expansion of the OMZ) in the CTBE interval (Jarvis et al., 1988; Jeans et al., 1991; Keller et al., 2001) while others prefer a scenario of decreasing productivity, with oceanic oligotrophic conditions spreading onto the shelf following a breakdown of the shelf-break fronts (zones of mixing that form barriers between shelf seas and open oceans), introducing pure chalk facies (Lamolda et al., 1994;

309

Gale et al., 2000). Prior to the CTBE, deposition in the central part of the Anglo-Paris Basin was dominated by marls and marly chalks, reflecting a significant input of terrigenous clay (Fig. 1A); afterwards oceanic white chalk deposition spread across the region (Fig. 1B) (Gale et al., 2000). Voigt et al. (2006) proposed increased riverine nutrient flux as an explanation for the first δ13C buildup (Plenus Marl Bed 1), and associated increased primary productivity and the development of deep sea anoxia, together with cooling induced by drawdown of atmospheric CO2, with Plenus Marl beds 2–6; they associated warming, transgression and the oligotrophication of European shelf seas with Plenus Marl beds 6–8 and above. There does seem to be good evidence that the OMZ expanded into the European shelf sea, affecting the plankton in the water column (e.g., Jarvis et al., 1988), but it is questionable whether it impinged on the sea floor and affected the benthos. According to Keller et al. (2001) the planktonic foraminifera show the first signs of an expanding OMZ impinging on the shelf sea at Eastbourne in Plenus Marl Bed 2, where it is marked by a decrease in surface dwelling taxa and an increase in heterohelicids which were tolerant of low oxygen levels. Fluctuations of proportions of planktonic foraminiferal taxa through the Plenus Marl are interpreted as marking fluctuations of productivity and oxygenation; increases of keeled taxa relative to heterohelicids in beds 3, 5 and 7 may signal a weakened OMZ and the development of water-mass stratification. Towards the top of the Plenus Marl (beds 7 and 8) a renewed increase in heterohelicids, the appearance of dwarfed taxa and a reduction of planktonic foraminiferal species diversity coincide with the second buildup of δ 13C; this is followed by dominance of heterohelicids and low species diversity in the overlying plateau phase of the δ 13C excursion, which Keller et al. (2001) considered to be indicative of high surface productivity and an expanded OMZ. As far as the benthos are concerned, however, the macrofaunal (including trace-fossil) and geochemical (pyritization levels) evidence argues for the persistence of oxygenated bottom waters in mid-shelf areas (e.g. Eastbourne) during the CTBE and does not support the idea of dysaerobia as an explanation of faunal turnover or local extinctions (Gale et al., 2000). The interpretation of the decline in the diversity of benthonic foraminifera and ostracods as being associated with dysaerobia (Jarvis et al., 1988) has also been challenged by Gale et al. (2000) who pointed out that it conflicts with the macrofossil and trace fossil evidence, and suggested a response to declining productivity as an alternative explanation. In fact the only evidence for bottom-water dysaerobia now seems to be that offered by the Platycopid Signal, the interpretation of which has itself been based on assumptions of bottom-water dysaerobia; the danger of a circular argument is all too apparent. An alternative explanation for the Platycopid Signal must be sought. Combining the interpretation of Gale et al. (2000) and Voigt et al. (2006) with the ostracod data, the following sequence of events can be envisaged: • Plenus Marl Bed 1—first buildup of δ 13C—influx of riverine nutrients following regresssion—increase in platycopid%. • Plenus Marl beds 2–6—peak of first buildup followed by δ 13C trough in beds 3–4, then second buildup—increased primary productivity (mesotrophic–eutrophic conditions), expansion of OMZ, deep ocean anoxia, drawdown of atmospheric CO2 leading to cooling—reduction of platycopid%, macrofaunal “cool pulse” in Bed 4 (Voigt et al., 2006) possibly reflected in the change in podocopid ostracod fauna (including increase of bairdioideans and cytheroideans). • Plenus Marl beds 7–8 and above—peak of second δ 13C buildup followed by plateau—warming, transgression, breakdown of shelfbreak fronts, spread of oligotrophic waters into shelf seas—major increase in platycopid%. This suggests that if the Platycopid Signal cannot be explained in terms of dissolved oxygen levels, an alternative might be found through consideration of trophic levels. Gebhardt and Zorn (2008) considered Cretaceous marine ostracods in an upwelling region (the

310

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

Fig. 4. Adult female paratypes of Cytherella rwhatleyi Brandão, 2008 (coll. nr ZMH K-41288): A, specimen SNB 0700, RV containing body and appendages (note egg in posterior brood space); B, detail of A showing overlapping mandibular and maxillular filter screens; C, specimen SNB 0698, detached pair of maxillular endopodites bearing filter screens of setulated setae (circle indicates example of typical, undisturbed setal spacing). Md = mandible, Mx = maxillula; arrows indicate anterior direction.

Tarfayah Basin, Morocco) to be influenced by water depth, oxygenation and food supply, remarking that the platycopid Cytherelloidea apparently avoided strong food pulses of high organic matter “rain” to the sea-floor. Puckett (1997) suggested that the overwhelming dominance of platycopids in the Campanian–Maastrichtian chalk of the US Gulf Coastal Plain might be partly due to the nature of the food supply which, although abundant, consisted of extremely fine-grained material which platycopids were better equipped than podocopids to utilise. These ideas are developed further in the following section. 6. An alternative Platycopid Signal Hypothesis In a system dominated by pelagic primary productivity (e.g., today's oceans; the Cretaceous Chalk shelf seas and oceans) the supply of “marine snow” or phytodetritus from surface waters to the sea floor is a major influence on the availability of food for benthos. Phytodetritus consists of aggregates of flocculated material from primary production in the photic zone, such as diatom blooms; it sinks rapidly to the ocean/sea floor before it can be eaten by pelagic zooplankton and thus constitutes a major food source for benthos; such aggregates typically contain a mixture of micro-, nano- and picoplankton which may include diatoms, coccolithophorids, dinoflagellates, silicoflagellates, phaeodarians, tintinnids, foraminifers, crustacean eggs and moults, protozoan faecal pellets, chlorophytes, cyanobacteria and bacteria, all embedded in a gelatinous matrix (Angel, 1990; Gooday and Turley, 1990; Shanks and Walters, 1997; Beaulieu and Smith, 1998; Iken et al., 2001; Turner, 2002). Deaths of larger nektonic forms including fish, squid and cetaceans may also

result in food packages arriving on the sea-floor. Plankton can be classified according to size as follows (from Fenchel, 1988): • mesoplankton, N200 μm (metazoans such as copepods) • microplankton, 20–200 μm (e.g. diatoms, coccolithophorids, dinoflagellates, radiolarians) • nanoplankton, 2–20 μm (e.g. dinoflagellates, chrysomonads, small diatoms and coccolithophorids) • picoplankton, 0.2–2.0 μm (e.g. bacteria, cyanobacteria) We have tried to estimate the particle sizes on which platycopids are likely to feed. The spacing between the main filter grille/screen setae on the mandibulae and maxillulae in Cytherella is 2–6 μm, but these spaces are crossed by setules, thus rendering the effective mesh size even smaller (Fig. 4). These figures are based on our measurements of a deep ocean platycopid, Cytherella rwhatleyi Brandão, 2008 (Fig. 7G and H of Brandão, 2008), and a shallow-water platycopid, Keijcyoidea infralittoralis Tsukagoshi, Okada and Horne, 2006 (Tsukagoshi et al., 2006, Figs. 7c and 8a). Cytherella rwhatleyi has 50–60 setae in the mandibular screen and 45–50 in the maxillular screen; in K. infralittoralis the screens have about half those numbers of setae but the intersetal spacing in both species is approximately the same. We acknowledge that these measurements, based as they are on illustrations of limbs that undoubtedly must have been disturbed and distorted (e.g., see Fig. 4c) during the process of dissection, are, at best, crude approximations. Moreover, the morphological characteristic that truly defines the mesh size is likely to be not the intersetal spacing but the intersetular spacing (as in other crustaceans—see, e.g., Fryer, 1983; Suh and Nemoto, 1987) which we are only able to estimate as being in the

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

order of 1–2 μm. Future detailed investigations of platycopid filter screen morphology (by means of scanning electron microscopy of material prepared by critical point drying) are likely to be instructive in this respect. Nevertheless, we submit that our measurements serve to give a first indication of the range of particle sizes on which platycopids are probably capable of feeding. Screens should be capable of catching particles N2 μm but given the overlap of two sets of screens with spaces 2–6 μm wide, coupled with the overlapping setules with spaces 1–2 μm wide, it is probable that particles as small as 1 μm, possibly even 0.5 μm, would be retained. The platycopid filter screen is thus adequate to collect nanoplankton, and given the cross-mesh provided by setules, probably at least the larger picoplankton. The majority of podocopids, as detritus/ deposit feeders with robust mandibular coxae for biting, tearing and chewing relatively large pieces of food, are not equipped with the filter screens necessary to feed efficiently on such fine suspended food particles. Where local hydrological conditions favour the concentration of such material, suspension feeders may become dominant (Angel, 1990). There are few data on living platycopid ontogeny, but Okada et al. (2008) have published detailed descriptions and illustrations of all the juvenile instars of Keijcyoidea infralittoralis, adults of which were described and illustrated by Tsukagoshi et al. (2006). It is noteworthy that in this species a fully-formed filter screen of about 20 setae first appears on the mandibula in the A-5 stage; the mandibula of the A-6 instar bears only five or six setae that might nevertheless fulfil the same function as the filter screen although their spacing does not seem particularly regular. The first two instars (A-7 and A-8) seem to lack any kind of filter screen on the mandibula. The maxillular filter screen appears for the first time in the A-4 instar, with about 20 setae. The setal width and spacing of the juvenile filter screens is essentially the same as that of the adults, and remains unchanged through ontogeny; in other words there is no proportional change in mesh size as the whole animal gets bigger. Only the number of setae increases, the mandibular screen from about 20 (A-5) to more than 30 (Adult) and the maxillular screen from about 20 (A-4) to more than 25 (Adult). This evidence seems to be consistent with the notion that platycopids are adapted to feed on a particular size range of particles and do so from the A-5 instar up to and including the adult stage; it is not clear how, or on what, the smaller juveniles (A-8 to A6) might feed. As Swanson et al. (2005) pointed out, platycopid feeding requires not only a supply of Particulate Organic Matter (POM) but that the particles should be of an appropriate size for their feeding appendages. They argued that coarser particles would need to be filtered out or blocked from entering the carapace (using this as an argument against filter-feeding). Pursuing this idea, it would seem probable that platycopids would do well under (1) very low energy conditions whereby only the finest particles (which they can feed on) remain in suspension—such as might obtain on the deep ocean floor, and (2) high energy conditions where the grain sizes that might clog their feeding appendages are winnowed away by currents, leaving only larger grains too big to enter the carapace gape—but where fine POM might be available in quiet periods (e.g., slack tide) or calm microenvironments such interstitial spaces in coarse shell sands or between algal fronds, both in shallow water. Mesotrophic and eutrophic conditions produce marine snow, which includes abundant microplankton. Oligotrophic conditions, on the other hand, are dominated by nano- and picoplankton, with bacterial biomass constituting a major component; photosynthetic picoplankton are important primary producers (Fenchel, 1988; Zohary and Robarts, 1992; Caron et al., 1999). The efficient nature of the microbial foodweb under such conditions can mean that little of the energy in the system becomes available as food to metazoan zooplankton (or, presumably, benthos) (Caron et al., 1999). It is easy to imagine deposit-feeding podocopid ostracods grazing on marine snow aggregates; moreover, being fairly mobile, like many

311

other benthos they can probably locate and exploit localised patches of phytodetritus as well as other food packages settling from the water column such as faecal pellets, carcasses and large plant remains (Angel, 1990; Gooday and Turley, 1990). However it seems unlikely that marine snow aggregates could be readily utilised by small meiobenthic filter-feeders such as platycopids; the particle size and aggregated nature of the food mean that it would not easily be sucked inside the carapace domicilium by the feeding current and any that did would probably clog the filter screens. It can be argued, accordingly, that mesotrophic and eutrophic conditions would favour podocopids, while oligotrophic conditions would favour platycopids. A further consideration is that an expanded OMZ in the water column would reduce zooplankton and lead to a reduction in the delivery, to the sea floor, of food in the form of metazoan faecal pellets. This would further reduce the availability of food to podocopids but not to platycopids; this means that in a sense the Platycopid Signal could indicate low oxygen levels, not in their ambient bottom waters but in the water column above. 7. Conclusion The existence of a Platycopid Signal (exceptionally high abundance of platycopids relative to podocopids) in ostracod assemblages from the CTBE in SE England is confirmed. The interpretation of the Platycopid Signal as indicative of low levels of dissolved oxygen in bottom waters is contradicted, however, by macrofossil, trace fossil and geochemical evidence which all imply a well-oxygenated seafloor throughout the CTBE. Evidence of the onset of oligotrophic conditions at the height of the CTBE coincides with the main Platycopid Signal. We propose the alternative hypothesis that assemblages overwhelmingly dominated by platycopids are primarily a signal of oligotrophy; this is supported by observations that living platycopids appear to be adapted to filter-feed on nano- and picoplankton which are the dominant groups in oligotrophic conditions. We consider this hypothesis to be applicable not only to the CTBE in the Anglo-Paris Basin but also potentially to all examples of the Platycopid Signal. Future studies may assess whether this alternative PSH can be supported or refuted, either by modern distributional and ecological data or by re-investigation of some of the other Palaeozoic and Mesozoic geological intervals in which a Platycopid Signal has been recognized. Acknowledgements DJH thanks Ian Jarvis for helpful discussions as well as for leading and coordinating sampling at Beachy Head, Eastbourne. We thank Holger Gebhardt and Marie-Béatrice Forel for their careful and thoughtful reviews which resulted in an improved final version of the paper. SNB thanks the Alexander von Humboldt Foundation for the fellowship. References Adamczak, F., 1969. On the question of whether the palaeocope ostracods were filter feeders. In: Neale, J.W. (Ed.), The Taxonomy, Morphology and Ecology of Recent Ostracoda. Oliver & Boyd, Edinburgh, pp. 93–98. Aiello, G., Barra, D., Bonaduce, G., Russo, A., 1996. The genus Cytherella Jones, 1849 (Ostracoda) in the Italian Tortonian-Recent. Revue de Micropaléontologie 39, 171–190. Angel, M.V., 1990. Food in the deep ocean. In: Whatley, R., Maybury, C. (Eds.), Ostracod and Global Events. British Micropalaeontological Society Publication Series. Chapman & Hall, London, pp. 274–285. Beaulieu, S.E., Smith, K.L., 1998. Phytodetritus entering the benthic boundary layer and aggregated on the sea floor in the abyssal NE Pacific: macro- and microscopic composition. Deep-Sea Research II 45, 781–815. Bergue, C.T., Coimbra, J.C., Cronin, T.M., 2007. Cytherellid species (Ostracoda), and their significance to the Late Quaternary events in the Santos Basin, Brazil. Senckenbergiana maritima 39, 5–12. Boomer, I., 1999. Late Cretaceous and Cainozoic bathyal Ostracoda from the Central Pacific (DSDP site 463). Marine Micropaleontology 37, 131–147.

312

D.J. Horne et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 308 (2011) 304–312

Boomer, I., Horne, D.J., Slipper, I.J., 2003. The use of ostracods in palaeoenvironmental studies, or what can you do with an ostracod shell. The Paleontological Society Papers 9, 153–179. Boomer, I., Whatley, R., 1992. Ostracoda and dysaerobia in the Lower Jurassic of Wales: the reconstruction of past oxygen levels. Palaeogeography, Palaeoclimatology, Palaeoecology 99, 373–379. Brandão, S.N., 2008. First record of a living Platycopida (Crustacea, Ostracoda) from Antarctic waters and a discussion on Cytherella serratula (Brady, 1880). Zootaxa 1866, 349–372. Brandão, S.N., Horne, D.J., 2009. The Platycopid Signal of oxygen depletion in the ocean: a critical evaluation of the evidence from modern ostracod biology, ecology and depth distribution. Palaeogeography, Palaeoclimatology, Palaeoecology 283, 126–133. Caron, D.A., Peele, E.R., Lim, E.L., Dennett, M., 1999. Picoplankton and nanoplankton and their trophic coupling in surface waters of the Sargasso Sea south of Bermuda. Limnology and Oceanography 44, 259–272. Crasquin-Soleau, S., Kershaw, S., 2005. Ostracod fauna from the Permian–Triassic boundary interval of South China (Huaying Mountains, eastern Sichuan Province): palaeoenvironmental significance. Palaeogeography, Palaeoclimatology, Palaeoecology 217 (1–2), 131–141. Fenchel, T., 1988. Marine plankton food chains. Annual Review of Ecology and Systematics 19, 19–38. Fryer, G., 1983. Functional ontogenetic changes in Branchinecta ferox (Milne-Edwards) (Crustacaea: Anostraca). Philosophical Transactions of the Royal Society, London B 303, 229–343. Gale, A.S., 1996. Turonian correlation and sequence stratigraphy of the Chalk in southern England. In: Hesselbo, S.B., Parkinson, D.N. (Eds.), Sequence Stratigraphy in British Geology, Geological Society Special Publication No. 103, pp. 177–195. Gale, A.S., Smith, A.B., Monks, N.E.A., Young, J.A., Howard, A., Wray, D.S., Huggett, J.M., 2000. Marine biodiversity through the Late Cenomanian–Early Turonian: palaeoceanographic controls and sequence stratigraphic biases. Journal of the Geological Society, London 157, 745–757. Gebhardt, H., 1999a. Cenomanian to Coniacian ostracodes from the Nkalagu area (SE Nigeria): biostratigraphy and palaeoecology. Paläontologische Zeitschrift 73, 77–98. Gebhardt, H., 1999b. Occurrence and palaeoecology of Cenomanian to Turonian ostracods from Ashaka (NE Nigeria). Neue Jahrbuch fur Geologie und Paläontologie Abhandlungen 211, 133–149. Gebhardt, H., Zorn, I., 2008. Cenomanian ostracods of the Tarfaya upwelling region (Morocco) as palaeoenvironmental indicators. Revue de Micropaléontologie 51, 273–286. Gooday, A.J., Turley, C.M., 1990. Responses by benthic organisms to inputs of organic material to the ocean floor: a review. Philosophical Transactions of the Royal Society, London A 331, 119–138. Iken, K., Brey, T., Wand, U., Voigt, J., Junghans, P., 2001. Food web structure of the benthic community at the Porcupine Abyssal Plain (NE Atlantic): a stable isotope analysis. Progress in Oceanography 50, 383–405. Horne, D.J., 2003. Key events in the ecological radiation of the Ostracoda. The Paleontological Society Papers 9, 181–201. Horne, D.J., 2005. Ostracoda. In: Selley, R.C., Cocks, R.M., Plimer, I.R. (Eds.), Encyclopaedia of Geology. Elsevier, Oxford, pp. 453–463. Horne, D.J., Jarvis, I., Rosenfeld, A., 1990. Recovering from the effects of an Ocean Anoxic Event: Turonian Ostracoda from S.E. England. In: Whatley, R.C., Maybury, C. (Eds.), Ostracod and Global Events. British Micropalaeontological Society Publication Series. Chapman & Hall, London, pp. 124–138. Horne, D.J., Slipper, I.J., 1992. Milankovitch cycles and microfossils: principals and practice of palaeocological illustrated by Cenomanian chalk-marl Rhythms by C.R. Paul—a comment. Journal of Micropalaeontology 11 (2), 241–242. Jarvis, I., Carson, G.A., Cooper, M.K.E., Hart, M.B., Leary, P.N., Tocher, B.A., Horne, D., Rosenfeld, A., 1988. Microfossil assemblages and the Cenomanian–Turonian (Late Cretaceous) oceanic anoxic event. Cretaceous Research 9, 3–103. Jarvis, I., Gale, A.S., Jenkyns, H.C., Pearce, M.A., 2006. Secular variation in Late Cretaceous carbon isotopes: a new δ13C carbonate reference curve for the Cenomanian– Campanian (99.6–70.6 Ma). Geological Magazine 43, 561–608. Jeans, C.V., Long, D., Hall, M.A., Bland, D.J., Cornford, C., 1991. The geochemistry of the Plenus Marls at Dover, England; evidence of fluctuating oceanographic conditions and of glacial control during the development of the Cenomanian–Turonian δ13C anomaly. Geological Magazine 128, 603–632. Johnson, N.A., 1996. The Cenomanian–Turonian Oceanic Anoxic Event: new ostracod data from Compton Bay, Isle of Wight, UK. In: Keen, M.C. (Ed.), Proceedings of the 2nd European Ostracodologists' Meeting, University of Glasgow, Scotland, 23 rd–27th July 1993. The British Micropalaeontological Society, London, pp. 43–48. Keller, G., Han, Q., Adatte, T., Burns, S.J., 2001. Palaeoenvironment of the Cenomanian– Turonian transition at Eastbourne, England. Cretaceous Research 22, 391–422.

Lamolda, M.A., Gorostidi, A., Paul, C.R.C., 1994. Quantitative estimates of calcareous nannofossil changes across the Plenus Marl (latest Cenomanian), Dover, England: implications for the interpretation of the Cenomanian–Turonian Boundary Event. Cretaceous Research 15, 143–164. Lethiers, F., Whatley, R.C., 1994. The use of Ostracoda to reconstruct the oxygen levels of Late Palaeozoic oceans. Marine Micropaleontology 24, 57–69. Lethiers, F., Whatley, R.C., 1995. Oxygénation des eaux et ostracodes filtreurs: application au Devonien–Dinantien. Geobios 28, 199–207. Majoran, S., Widmark, J.G.V., Kucera, M., 1997. Palaeoecological preferences and geographical distribution of Late Maastrichtian deep-sea ostracods in the South Atlantic. Lethaia 30, 53–64. Majoran, S., Widmark, J.G.V., 1998. Response of deep-sea ostracod assemblages to Late Cretaceous palaeoceanographical changes: ODP Site 689 in the Southern Ocean. Cretaceous Research 19, 843–872. Mortimore, R.N., Wood, C.J., Gallois, R.W., 2001. British Upper Cretaceous Stratigraphy. Joint Nature Conservation Committee, Peterborough: Geological Conservation Review Series, No. 23. 558 pp. Okada, R., Tsukagoshi, A., Smith, R.J., Horne, D.J., 2008. The ontogeny of the platycopid Keijcyoidea infralittoralis (Ostracod: Podocopa). Zoological Journal of the Linnean Society 153, 213–237. Olempska, E., 2008. Soft body-related features of the carapace and the lifestyle of Paleozoic beyrichioidean ostracods. Journal of Paleontology 82, 717–736. Paul, C.R.C., Lamolda, M.A., Mitchell, S.F., Vaziri, M.R., Gorostidi, A., Marshall, J.D., 1999. The Cenomanian–Turonian boundary at Eastbourne (Sussex, UK): a proposed European reference section. Palaeogeography, Palaeoclimatology, Palaeoecology 150, 83–121. Puckett, T.M., 1997. The genus Krithe (Ostracoda) from the Campanian and Maastrichtian (Upper Cretaceous) of the northern US Gulf Coastal Plain. Journal of Micropalaeontology 16, 145–157. Shanks, A.L., Walters, K., 1997. Holoplankton, meroplankton, and meiofauna associated with marine snow. Marine Ecology Progress Series 156, 75–86. Slipper, I.J., 1996. Early Turonian Ostracoda: the Melbourn Rock fauna from Abbots Cliff, Dover, England. In: Keen, M.C. (Ed.), Proceedings of the 2nd European Ostracodologists' Meeting, University of Glasgow, Scotland, 23 rd–27th July 1993. The British Micropalaeontological Society, London, pp. 49–56. Smith, A.J., Horne, D.J., 2002. Ecology of marine, marginal marine and nonmarine ostracodes. In: Holmes, J.A., Chivas, A.R. (Eds.), The Ostracoda: Applications in Quaternary research, American Geophysical Union. Geophysical Monograph Series 131, 37–64. Steineck, P.L., Thomas, E., 1996. The latest Paleocene crisis in the deep sea: Ostracode succession at Maud Rise, Southern Ocean. Geology 24, 583–586. Suh, H.-L., Nemoto, T., 1987. Comparative morphology of filtering structure of five species of Euphausia (Euphausiacea, Crustacea) from the Antarctic Ocean. Proceedings of the NIPR Symposium on Polar Biology, vol. 1, pp. 72–83. Swanson, K.M., Jellinek, T., Malz, H., 2005. The platycopine condition: new observations on reproduction, respiration and feeding, in living, deep-sea Platycopina (Crustacea, Ostracoda). Senckenbergiana maritima 35, 157–187. Tsukagoshi, A., Okada, R., Horne, D.J., 2006. Appendage homologies and the first record of eyes in platycopid ostracods, with the description of a new species of Keijcyoidea (Crustacea: Ostracoda) from Japan. Hydrobiologia 559, 255–274. Turner, J.T., 2002. Zooplankton fecal pellets, marine snow and sinking phytoplankton blooms. Aquatic Microbial Ecology 27, 57–102. Voigt, S., Gale, A.S., Voigt, T., 2006. Sea-level change, carbon cycling and palaeoclimate during the Late Cenomanian of northwest Europe; an integrated palaeoenvironmental analysis. Cretaceous Research 27, 836–858. Weaver, P.P.E., 1981. The distribution of Ostracoda in the British Cenomanian. In: Neale, J.W., Brasier, M.D. (Eds.), Microfossils from Recent and Fossil Shelf Seas. Ellis Horwood, Chichester, pp. 156–162. Whatley, R.C., 1990. Ostracoda and global events. In: Whatley, R.C., Maybury, C. (Eds.), Ostracoda and Global Events. British Micropalaeontological Society Publication Series. Chapman & Hall, London, pp. 3–24. Whatley, R.C., 1991. The Platycopid Signal: a means of detecting kenoxic events using Ostracoda. Journal of Micropalaeontology 10 (2), 181–185. Whatley, R.C., 1995. Ostracoda and oceanic palaeoxygen levels. Mitteilungen aus dem Hamburgischen Zoologischen Museum und Institut 92, 337–353. Whatley, R.C., Arias, C.F., 1993. Palaeogene Ostracoda from the Tripoli Basin Libya. Revista Española Micropaleontología 25, 125–154. Whatley, R.C., Arias, C.F., Comas-Rengifo, M.J., 1994. The use of Ostracoda to detect kenoxic events: a case history from the Spanish Toarcian. Geobios 17, 733–741. Whatley, R.C., Pyne, R.S., Wilkinson, I.P., 2003. Ostracoda and palaeo-oxygen levels, with particular reference to the Upper Cretaceous of East Anglia. Palaeogeography, Palaeoclimatology, Palaeoecology 194, 355–386. Zohary, T., Robarts, R.D., 1992. Bacterial numbers, bacterial production, and heterotrophic nanoplankton abundance in a warm core eddy in the Eastern Mediterranean. Marine Ecology Progress Series 84, 133–137.