Mid-Pleistocene benthic foraminiferal record in the Southwest Pacific

Benthic foraminiferal faunas from three bathyal sequences provide a proxy record of oceanographic changes through the mid-Pleistocene transition (MPT) on either side of the Subtropical Front (STF), east of New Zealand. Canonical correspondence analyses show that factors related to water depth, latitude and climate cycles were more significant than oceanographic factors in determining changes in faunal assemblage composition over the last 1 Ma. Even so, mid-Pleistocene faunal changes are recognizable and can be linked to inferred palaeoceanographic causes. North of the largely stationary STF the faunas were less variable than to the south, perhaps reflecting the less extreme glacial-interglacial fluctuations in the overlying Subtropical Surface Water. Prior to Marine Isotope Stage (MIS) 21 and after MIS 15, the northern faunas had fairly constant composition, but during most of the MPT faunal composition fluctuated in response to climate-related food-supply variations. Faunal changes through the MPT suggest increasing food supply and decreasing dissolved bottom oxygen. South of the STF, beneath Subantarctic Surface Water, mid-Pleistocene faunas exhibited strong glacial-interglacial fluctuations, inferred to be due to higher interglacial nutrient supply and lower oxygen levels. The most dramatic faunal change in the south occurred at the end of the MPT (MIS 17- 12). with an acme of Abditodentrix pseudothalmanni, possibly reflecting higher carbon flux and lower bottom oxygen. This study suggests that the mid-Pleistocene decline and extinction of a group of elongate, cylindrical deep-sea foraminifera may have been related to decreased bottom oxygen concentrations as aresult of slower deep-water currents.

Benthic foraminifera in surface sediments of the Gulf of Lions

Surface sediment was sampled at two bathyal sites in the southwestern Gulf of Lions in the western Mediterranean Sea in February and August 1997 to study the distribution and microhabitat of living (Rose Bengal stained) deep sea benthic foraminifera. Both standing stock and diversity of the faunas, and the microhabitat of distinct species mirror the trophic situation and the depth of the oxidised layer at the different sites. Our results suggest that the faunas do not comprise highly opportunistic species and are adapted to rather stable environments. In the axial channel of the Lacaze-Duthiers Canyon, organic matter fluxes are enhanced due to advective transport of organic matter resulting in elevated oxygen consumption rates in the surface sediment and a rather thin oxidised layer. The corresponding benthic foraminiferal fauna is characterised by rather high standing stock and diversity, and a well-developed deep infauna. In addition to freshly deposited phytodetritus, more degraded organic matter seems to be an important food source. In contrast, at the open slope, organic matter fluxes and oxygen consumption rates in the surface sediment are lower and the oxidised layer is much thicker than inside the canyon. The corresponding benthic foraminiferal fauna comprises mainly epifaunal and shallow-infaunal species with much lower standing stocks and clear differences between February and August. In August standing stocks are higher and the average living depths of most species shift towards the sediment surface. These differences can be attributed to patchiness or represent a seasonal trophic signal.

Miozene larger foraminifera in ODP Leg 133 holes

The Marion Plateau is a large carbonate platform off northeastern Queensland. Three sites (815, 816, and 826) were drilled on this platform and form the basis for this study. Larger benthic foraminifers, together with rare planktonic forms from the shallow-water carbonates that form the main part of the platform sequence, were studied to establish a biostratigraphy. The presence of Lepidocyclina (Nephrolepidiná) howchini sensu lato and Ladoronia vermicularis, together with Globorotalia (Globorotalia) praemenardii and Orbulina, indicate an early middle Miocene (N9-N12) age (i.e., lower Tf stage) for these carbonates. Dolomitization has destroyed much of the original fabric of these carbonates, making study of the larger foraminifers difficult. Sites 815 (forereef location) and 826 (backreef, lagoonal setting) provide the best faunas. However, at all sites nodular coralline algae and Halimeda are the major bioclasts; coral fragments form a major component at Sites 816 and 826. The middle Miocene neritic sequence is separated from the overlying hemipelagic sequence by an unconformity that spans much of the middle and late Miocene. At Site 815, which is in a forereef situation, the overlying hemipelagic sequence contains a Zone N17A fauna, but at Site 816, higher on the platform, a similar sequence contains a Zone N19 fauna. The faunas indicate that the platform was built up during the early middle Miocene and remained at fairly constant water depths and temperatures during this period. It was then exposed prior to subsiding rapidly during the late Miocene and Pliocene to depths similar to those of the present day.

Planktic foraminifera across the Cretaceous-Tertiary transition in the Antarctic Ocean

Three Antarctic Ocean K/T boundary sequences from ODP Site 738C on the Kerguelen Plateau, ODP Site, 752B on Broken Ridge and ODP Site 690C on Maud Rise, Weddell Sea, have been analyzed for stratigraphic completeness and faunal turnover based on quantitative planktic foraminiferal studies. Results show that Site 738C, which has a laminated clay layer spanning the K/T boundary, is biostratigraphically complete with the earliest Tertiary Zones P0 and P1a present, but with short intrazonal hiatuses. Site 752B may be biostratigraphically complete and Site 690C has a hiatus at the K/T boundary with Zones P0 and P1a missing. Latest Cretaceous to earliest Tertiary planktic foraminiferal faunas from the Antarctic Ocean are cosmopolitan and similar to coeval faunas dominating in low, middle and northern high latitudes, although a few endemic species are present. This allows application of the current low and middle latitude zonation to Antarctic K/T boundary sequences. The most abundant endemic species is Chiloguembelina waiparaensis, which was believed to have evolved in the early Tertiary, but which apparently evolved as early as Chron 30N at Site 738C. Since this species is only rare in sediments of Site 690C in the Weddell Sea, this suggests that a watermass oceanographic barner may have existed between the Indian and Atlantic Antarctic Oceans. The cosmopolitan nature of the dominant fauna began during the last 200,000 to 300,000 years of the Cretaceous and continued at least 300,000 years into the Tertiary. This indicates a long-term environmental crisis that led to gradual elimination of specialized forms and takeover by generalists tolerant of wide ranging temperature, oxygen, salinity and nutrient conditions. A few thousand years before the K/T boundary these generalists gradually declined in abundance and species became generally dwarfed due to increased environmental stress. There is no evidence of a sudden mass killing of the Cretaceous fauna associated with a bolide impact at the K/T boundary. Instead, the already declining Cretaceous taxa gradually disappear in the early Danian and the opportunistic survivor taxa (Ch. waiparaensis and Guembelitria cretacea) increase in relative abundance coincident with the evolution of the first new Tertiary species.

Middle Eocene to early Oligocene planktonic and benthic foraminifers from ODP Hole 125-786A

Drilling at Site 786, located in the center of the Izu-Bonin forearc basin, penetrated an apparently continuous section of middle Eocene/lower Oligocene volcaniclastic breccias and nannofossil oozes. Planktonic foraminiferal faunas underwent a gradual transition from relatively high-diversity middle Eocene through late Eocene tropical or warm-water assemblages to a cooler-water, less diverse assemblage during the early Oligocene. In the cosmopolitan benthic foraminiferal faunas, the major transition occurred during the early late Eocene. Middle Eocene benthic assemblages resembling the bathyal 'Lenticulina' fauna (characterized by Osangularia mexicana, Cibicidoides eocaenus, and several buliminid species) changed to an upper Eocene abyssal 'Globocassidulina subglobosa' fauna (characterized by Cibicidoides praemundulus, Globocassidulina subglobosa, Gyroidinoides girardanus, Oridorsalis umbonatus, and Siphonodosaria aculeata). Even though no large, abrupt faunal changes appear to have been associated with the assumed Eocene/Oligocene boundary, benthic species turnover continued through the late Eocene and into the early Oligocene. This resulted in a slightly lower diversity early Oligocene fauna dominated by three species: Laevidentalina sp., Bulimina jarvisi, and Gyroidinoides girardanus. The progression from a middle Eocene bathyal 'Lenticulina' fauna, rather than an abyssal 'Nuttallides truempyi' fauna, to an abyssal 'Globocassidulina subglobosa' fauna during the early late Eocene, suggests that a bathymetric deepening occurred at Site 786. Increased water depths may have resulted from tectonic subsidence.

Neogene and Quaternary paleoenvironmental history at DSDP Leg 71 Holes

The Neogene and Quaternary sedimentary record of Leg 71 and previously drilled sequences from the Southern Ocean reveal evidence of a major late Miocene change of oceanic and glacial conditions in the southern high latitudes during paleomagnetic Chron 9. The characteristics of late Miocene sedimentation and in particular the study of erosional patterns and ice-rafted debris suggest the following conclusions. 1) In the late Miocene, the Polar Front first migrated to the northern latitudes of the Southern Ocean and surface water temperatures became similar to those of today. 2) Extensive ice shelves or ice tongues were not present along the Antarctic margin until late Chron 9 (about 9.0 Ma). 3) Before Chron 9, West Antarctica was occupied by an archipelago and the West Antarctic Sea. 4) Extensive ice shelves formed in the West Antarctic region, eventually coalescing and thickening to form the grounded West Antarctic ice sheet by Chron 9. 5) The newly formed West Antarctic ice sheet was probably unstable and frequently became an ungrounded ice shelf, thus accounting for the scarcity of late Miocene ice-rafted debris. 6) Extensive erosion or nondeposition of sediment was probably the result of increased Antarctic Bottom Water (AABW) formation in the West Antarctic region during the initial formation of extensive West Antarctic ice shelves and during periods when the West Antarctic ice sheet was ungrounded. 7) In the Southwest Atlantic, AABW velocity waned during the latest Miocene. During the late Gilbert Chron a major and permanent change occurred in the pattern of ice-rafting to the South Atlantic, and after 4.35 Ma the increased IRD accumulation rate and frequency of major episodes of IRD accumulation suggest increased stability of the West Antarctic ice sheet. In addition, radiolarian faunas of Hole 514 record at least eight migrations of the Polar Front to the north of the site during the past 4.07 m.y. An apparent increase in the frequency of Polar Front migrations occurred about 2.7-2.6 Ma, possibly in response to oceanic change induced by fluctuations in glacial conditions of the Northern Hemisphere.

Miocene to Pleistocene planktonic foraminifer biostratigraphy of ODP Leg 135 sites

Diverse and well-preserved planktonic foraminifers were recovered from six sites (834-839) drilled in the Lau Basin. Planktonic faunas from the Tongan Platform sites varied from those of the Lau Basin sites by being less well preserved (Site 840) to being very poorly preserved and very sparse (Site 841); at Site 841 most samples were barren. All sites penetrated a volcaniclastic sequence in which thick ash beds were encountered; foraminifer populations within the ash beds were often very small, making it difficult to obtain biostratigraphic data. No hiatuses were encountered in the upper Miocene to Pleistocene sections of the Lau Basin, but a possible break occurs at Site 840 on the Tongan Platform. Site 834 penetrated through a Quaternary-Pliocene sequence overlying basaltic basement, and topmost Miocene (Zone N17B) sediments interbedded within the volcanic sequence. Site 835 penetrated into the lower Pliocene (Zones N19 to N19-20). Site 836 penetrated the shortest section, with Zone N22 {Globorotalia (Truncorotalia) crassaformis hessi Subzone) directly overlying basalts. Site 837 penetrated into the basal part of Zone N22 (Globigerinoides quadrilobatus fistulosus Subzone) overlying basalt. Site 838 failed to encounter basalts, with the oldest sediment being from Zone N22 (Globigerinoides quadrilobatus fistulosus Subzone). Site 839, within the same basin as Site 838, located Zone N22 (Globigerinoides quadrilobatus fistulosus Subzone) sediments directly overlying igneous basement. Site 840 penetrated into the upper Miocene Zone N17A without encountering any major unconformity. Site 841, studied mainly from core-catcher samples, penetrated a Quaternary to questionable upper Miocene sequence that was in fault contact with middle Miocene (Zones N8 to N9) sediments. For the Lau Basin sites, reworking was encountered only in Sites 834 and 835. Site 834 was drilled adjacent to the Lau Ridge, on which are developed numerous reef al and shallow-water environments, where erosional conditions could have been expected during sea-level lowstands. Site 835 was drilled in a narrow basin that has been remote from these erosional influences; slumping and erosion of material from the adjacent basin slopes appears to have been the source of the reworking. For the Tongan Platform sites, reworking was observed only in the lower part of the upper Miocene section at Site 841, where late Eocene larger foraminifers are present in conglomerates and grits. The presence of Globorotalia (Globorotalia) multicamerata and small specimens of Sphaeroidinellopsis spp. in the Pleistocene of Site 840 may indicate reworking, but this is not clear. Unit I, which marks a reduction in volcanic activity in the Lau Basin, ranges in age from the lower part of Zone N22 (Globigerinoides quadrilobatus fistulosus Subzone) at Sites 834 and 835, to within Zone N22 (Globorotalia crassaformis hessi Subzone) at Sites 836 to 838, and within the upper part of Zone N22 (Bolliella praeadamsi Subzone) at Site 839. Units II and III are generally represented by thick to very thick ash beds, which generally contain low-diversity and often poorly preserved assemblages. Igneous sources seem to have remained important contributors of sediment up to the present day.

Current velocities and settling experimental results

This study provides a theoretical assessment of the potential bias due to differential lateral transport on multi-proxy studies based on a range of marine microfossils. Microfossils preserved in marine sediments are at the centre of numerous proxies for paleoenvironmental reconstructions. The precision of proxies is based on the assumption that they accurately represent the overlying watercolumn properties and faunas. Here we assess the possibility of a syn-depositional bias in sediment assemblages caused by horizontal drift in the water column, due to differential settling velocities of sedimenting particles based on their shape, size and density, and due to differences in current velocities. Specifically we calculate the post-mortem lateral transport undergone by planktic foraminifera and a range of other biological proxy carriers (diatoms, radiolaria and fecal pellets transporting coccolithophores) in several regions with high current velocities. We find that lateral transport of different planktic foraminiferal species is minimal due to high settling velocities. No significant shape- or size-dependent sorting occurs before reaching the sediment, making planktic foraminiferal ideal proxy carriers. In contrast, diatoms, radiolaria and fecal pellets can be transported up to 500 km in some areas. For example in the Agulhas current, transport can lead to differences of up to 2°C in temperature reconstructions between different proxies in response to settling velocities. Therefore, sediment samples are likely to contain different proportions of local and imported particles, decreasing the precision of proxies based on these groups and the accuracy of the temperature reconstruction.

Cenozoic plaktonic foraminfers from the Celebes and Sulu Seas

This paper presents the planktonic foraminifer biostratigraphy of the sites drilled during Ocean Drilling Program Leg 124 in the Celebes and Sulu Seas. It discusses preservation of foraminifers in pelagic sediments and in calcareous turbidites. In the Celebes Sea, pelagic carbonates are only found in the Eocene and Oligocene at Site 770. The faunas are poorly preserved due to severe dissolution and offer little biostratigraphic detail. In the Sulu Sea, pelagic carbonates are found in the upper Pliocene and Pleistocene at Sites 768 and 769 and throughout the recovered sequence at the shallower Site 771. The foraminifer faunas from these sediments allow for recognition of most standard zones. Variations in preservation of pelagic foraminifer faunas with time are due to changes in the depth of the lysocline. Shifts to improved preservation at Sites 768 and 769 are synchronous in the upper Pliocene/Pleistocene and may be related to global sea-level cycles. Planktonic foraminifers are also abundant in calcareous turbidites, which were deposited in both basins from the late Miocene onward. However, the turbidites are fine-grained, and biostratigraphic marker species are absent as a result of size-sorting during transport. In the Celebes Sea, shelf-derived material was a major component of early-late Miocene and middle Pliocene to early Pleistocene turbidites. Changes in the composition of the turbidites may correspond to global sea-level changes. In the Sulu Sea, a shift from shelf-derived material in Pliocene calcareous turbidites to a pelagic source in the Pleistocene may be related to subsidence of the Cagayan Ridge.

Mid-Pleistocene planktonic foraminiferal composition and IRD content in ODP Sites 162-980 and 162-984

We investigated surface and deep ocean variability in the subpolar North Atlantic from 1000 to 500 thousand years ago (ka) based on two Ocean Drilling Program (ODP) sites, Feni drift site 980 (55°29'N, 14°42'W) and Bjorn drift site 984 (61°25'N, 24°04'W). Benthic foraminiferal stable isotope data, planktic foraminiferal faunas, ice-rafted debris data, and faunally based sea-surface temperature estimates help test the hypothesis that oceanographic changes in the North Atlantic region were associated with the onset of the 100-kyr world during the mid-Pleistocene revolution. Based on percentage of Neogloboquadrina pachyderma (s) records from both sites, surface waters during interglacials and glacials were cooler in the mid-Pleistocene than during marine isotope stages (MIS) 5 and 6. In particular, interglaciations at Bjorn drift site 984 were significantly cooler. Faunal evidence suggests that the interglacial Arctic front shifted from a position between the two sites to a position northwest of Bjorn drift site 984 after ca. 610 ka. As during the late Pleistocene, we find faunal evidence for lagging surface warmth at most of the glacial initiations during the mid-Pleistocene. Each initiation is associated with high benthic d13C values that are maintained into the succeeding glaciation, which we term "lagging NADW production." These findings indicate that lagging warmth and lagging NADW production are robust features of the regional climate system that persist in the middle to late Pleistocene.

Jurassic microfossils of DSDP Hole 79-547B

The first marine incursion of the incipient North Atlantic Ocean is recorded in the uppermost Triassic to Lower Jurassic sequence of DSDP Site 547 off central Morocco. A lithologic change from continental red beds below to slope breccias and hemipelagic carbonates above indicates that a carbonate ramp was probably established by Sinemurian time along the Moroccan continental margin and that subsidence in the adjacent basin was rapid in the early phases of continental rift. Foraminifers recovered from the Liassic (Sinemurian-Pliensbachian) basinal deposits are diverse and well preserved. The faunas are compositionally similar to contemporaneous neritic assemblages of Europe and the Grand Banks of Newfoundland. The Middle Jurassic in Hole 547B is characterized by regressive deposits that are poor in foraminifers. The major Late Jurassic "Atlantic" transgression is again represented by basinal deposits consisting of limestone breccias and pelagic carbonates. Foraminifers recovered from this interval are transitional between Late Jurassic assemblages reported from deep-sea deposits in the North Atlantic and typical Late Jurassic neritic assemblages of Europe. The Late Jurassic assemblages of Hole 547B are primarily dominated by nodosariids and spirillinids with moderate abundances of simple arenaceous forms. Nonreticulate epistominids occur very rarely in the Upper Jurassic of Hole 547B. It is tentatively suggested that these represent upper bathyal assemblages.

Neptunian dikes and their fillings in carbonates of the Warstein area (northeastern Rhenish massif)

Neptunian dikes and cavities as weil as their fillings are described from Middle to Upper Devonian carbonates of the Warstein area. The genesis of the pre-Upper Carboniferous dikes is due to pre-orogenic synsedimentary tensional movements. Lifting, subsidence and tilting caused joints and cracks, which are enlarged to dikes and cavities on submarine conditions. The post-Upper Carboniferous dikes are based on the orogenesis during Upper Carboniferous time, causing numerous tectonical divisional planes in the sediments. Along these planes a far-reaching karstification took place since mesozoic time. According to their size the cavities are subdivided into macro-, mega- and microdikes. With the exception of one macrodike all the others are limited to the massive limestone. Megadikes especially occur in Upper Devonian cephalopod limestone and in the Erdbach limestone, microdikes can be found in all carbonatic rocks. The dikes follow pre-orogenic, tectonical and sedimentary divisional planes and are orientated to ac-, bc- as well as bedding planes and diagonal directions. The fillings happened down from above either in a solitary event or repeatedly in long-lived dikes during a span of several ten millions of years. More seldom the fillings took place laterally or upside from beneath. The dikes contain - without regard to autochthonous conodont faunas - older and/or younger mixed faunas, too. Occasionally they were used as life district by a trilobite fauna adapted to the dikes. The dikes represent sedimentary pitfalls and conserve sediments eroded in other places. Therefore, by aid of the fillings, it can be demonstrated, that stratigraphic gaps are not absolutely due to primary interruptions of sedimentation, but were caused by reworking. Some dikes contain the distal offsets of slides and suspension streams. Relations between condensation and development of dikes could not be derived in the Warstein area. However, an increase of the frequency of dikes towards east to the eastern margin of the Warstein carbonate platform could be pointed out. This margin is a slope, persisting more than 10 millions of years, between a block and a basin. Evidently cracks and dikes, which were caused by settlements, slides and earth quakes, occured there frequently. The Warstein dikes and cavities, caused by karstification, are filled with terrestrial Lower Cretaceous, marine Upper Cretaceous and terrestrial Pleistocene to Holocene sediments. Tertiary sediments could not be detected.

Neogen carbonate and stable isotope record of ODP Hole 120-751A

Lower Miocene through upper Pleistocene benthic foraminifer assemblage records from Ocean Drilling Program Site 751 on the Southern Kerguelen Plateau (57°44'S, water depth 1634 m) were combined with benthic and planktonic foraminifer oxygen and carbon isotope records and high-resolution CaCO3 data from the same site. Implications for the Neogene productivity and paleoceanography of the southern Indian Ocean are discussed. We used distinctive features of the Miocene d18O and d13C curves for stratigraphic correlation. Coinciding with a lower middle Miocene hiatus from 14.2 to 13.4 Ma, there was a rapid increase in benthic d18O values by 1.2 per mil. This distinct increase occurs in middle Miocene benthic foraminifer oxygen isotope curves from all oceans. No major change, however, in benthic foraminifer faunal composition occurred in this period of growth of the Antarctic ice cap and cooling of deep ocean waters (14.9-14.2 Ma). A drastic change in benthic foraminifer faunas coincided with a hiatus from 8.4 to 5.9 Ma. Shortly after this hiatus, in the latest Miocene, the CaCO3 content of the sediments dropped from 75% to 0%. From that time ( 5.8 Ma) through the early Pliocene, Site 751 has been situated beneath a high biogenic siliceous productivity zone. Carbonate contents of upper Pliocene and Pleistocene sediments vary between 20% and 70%. The benthic foraminifer faunas in the uppermost Pliocene and lower Pleistocene reflect strong bottom current conditions, in contrast to those in the upper Pleistocene, which indicate calm sedimentation and high food supply. High d13C values of planktonic foraminifers compared with low values of benthic foraminifers suggest high primary productivity in the late Pleistocene. The changes in productivity were probably a result of latitudinal migration and meandering of the Polar Frontal Zone.

Distribution of deep-sea foraminifera in surface sediments east of New Zealand

Paleobathymetric assessments of fossil foraminiferal faunas play a significant role in the analysis of the paleogeographic, sedimentary, and tectonic histories of New Zealand's Neogene marine sedimentary basins. At depths >100 m, these assessments often have large uncertainties. This study, aimed at improving the precision of paleodepth assessments, documents the present-day distribution of deep-sea foraminifera (>63 µm) in 66 samples of seafloor sediment at 90-700 m water depth (outer shelf to mid-abyssal), east of New Zealand. One hundred and thirty-nine of the 465 recorded species of benthic foraminifera are new records for the New Zealand region. Characters of the foraminiferal faunas which appear to provide the most useful information for estimating paleobathymetry are, in decreasing order of reliability: relative abundance of common benthic species; benthic species associations; upper depth limits of key benthic species; and relative abundance of planktic foraminifera. R mode cluster analysis on the quantitative census data of the 58 most abundant species of benthic foraminifera produced six species associations within three higher level clusters: (1) calcareous species most abundant at mid-bathyal to outer shelf depths (<1000 m); (2) calcareous species most abundant at mid-bathyal and greater depths (>600 m); (3) agglutinated species mostly occurring at deep abyssal depths (>3000 m). A detrended correspondence analysis ordination plot exhibits a strong relationship between these species associations and bathymetry. This is manifest in the bathymetric ranges of the relative abundance peaks of many of the common benthic species (e.g., Abditodentrix pseudothalmanni 500-2800 m, Bolivina robusta 200-650 m, Bulimina marginata f. marginata 20-600 m, B. marginata f. aculeata 400-3000 m, Cassidulina norvangi 1000-4500 m, Epistominella exigua 1000-4700 m, and Trifarina angulosa 10-650 m), which should prove useful in paleobathymetric estimates. The upper depth limits of 28 benthic foraminiferal species (e.g., Fursenkoina complanata 200 m, Bulimina truncana 450 m, Melonis affinis 550 m, Eggerella bradyi 750 m, and Cassidulina norvangi 1000 m) have potential to improve the precision of paleobathymetric estimates based initially on the total faunal composition. The planktic percentage of foraminiferal tests increases from outer shelf to upper abyssal depths followed by a rapid decline within the foraminiferal lysocline (below c. 3600 m). A planktic percentage <50% is suggestive of shelf depths, and >50% is suggestive of bathyal or abyssal depths above the CCD. In the abyssal zone there is dramatic taphonomic loss of most agglutinated tests (except some textulariids) at burial depths of 0.1-0.2 m, which negates the potential usefulness of these taxa in paleobathymetric assessments.

Measurements of gastropods shell sizes from 23 fossil and extant long-lived lakes

Aim: To investigate shell size variation among gastropod faunas of fossil and recent long-lived European lakes and discuss potential underlying processes. Location: 23 long-lived lakes of the Miocene to Recent of Europe. Methods: Based on a dataset of 1412 species of both fossil and extant lacustrine gastropods, we assessed differences in shell size in terms of characteristics of the faunas (species richness, degree of endemism, differences in family composition) and the lakes (surface area, latitude and longitude of lake centroid, distance to closest neighbouring lake) using multiple and linear regression models. Because of a strong species-area relationship, we used resampling to determine whether any observed correlation is driven by that relationship. Results: The regression models indicated size range expansion rather than unidirectional increase or decrease as the dominant pattern of size evolution. The multiple regression models for size range and maximum and minimum size were statistically significant, while the model with mean size was not. Individual contributions and linear regressions indicated species richness and lake surface area as best predictors for size changes. Resampling analysis revealed no significant effects of species richness on the observed patterns. The correlations are comparable across families of different size classes, suggesting a general pattern. Main conclusions: Among the chosen variables, species richness and lake surface area are the most robust predictors of shell size in long-lived lake gastropods. Although the most outstanding and attractive examples for size evolution in lacustrine gastropods derive from lakes with extensive durations, shell size appears to be independent of the duration of the lake as well as longevity of a species. The analogue of long-lived lakes as 'evolutionary islands' does not hold for developments of shell size because different sets of parameters predict size changes.