Eocene-Pliocene deep sea ostracodes from ODP site 744A, Southern Indian Ocean

The Eocene-Pliocene deep sea ostracodes from the ODP site 744A (Kerguelen Plateau) are herein studied under the taxonomic and paleoecologic aspects. 28 species are identified, being the genera Krithe, Cytherella and Dutoitella the most diversified. A faunal threshold was recorded in the Early Oligocene, which is tentatively explained under the knowledge of the paleoceanographical studies carried out not only in the Kerguelen Plateau but also in adjacent areas. The faunal turnover and variations in both richness and abundance possibly reflect the inception of psychrosphere and the influence of hydrological changes in the preservation of carapaces. Moreover, the influence of those changes on carbonate preservation is discussed as the cause of faunal impoverishment in the upper portion of the core.


INTRODUCTION
Deep sea ostracode research has developed significantly in the last few decades, with improved taxonomic, ecologic and zoogeographic information.Studies carried out from 1970 onwards (see Benson 1988 andCronin et al. 2002 revisions) brought evidence that continental slopes and oceanic basins are inhabited by well-diversified and distinct faunas.The distribution, diversity and abundance of slope and abyssal plain assemblages are strongly influenced by the local hydrologic structure as well as climatic driven oceanographic events, even over short timescales (Ayress et al. 1997, Yasuhara et al. 2008).
The Cenozoic deep sea ostracodes have their origin from Late Cretaceous shallow water stocks (Benson 1975).According to this author, a worldwide faunal change at approximately 40 Ma established an oceanic psychrosphere, which influenced the evolution of fau-748 CRISTIANINI T. BERGUE and ABIRAMAN GOVINDAN Kerguelen Plateau, is uniquely positioned to record the climatic evolution of the Southern Ocean region and its hydrologic changes.Studies on the cored material of this site include Huber (1999) on planktonic foraminiferal biozonation, Schröder-Adams (1991) on benthic foraminifera, Caulet (1991) on radiolarian biostratigraphy, and Baldauf and Barron (1991) on diatom correlation.Details of the lithostratigraphy of this site are outlined in Barron et al. (1991).However, the Paleogene and Neogene ostracodes from this site have not been studied so far.
Recent ostracodes from the Southern Ocean have been fairly well documented since the pioneering study of Brady (1880) as reviewed by Ayress et al. (2004).Some studies on Paleogene and Neogene assemblages have also been published, such as Guernet (1985), Guernet and Galbrun (1992) and Steineck and Thomas (1996).The main objective of this article is to present a preliminary study on the ostracode fauna of Late Eocene to Pliocene from site 744A as a contribution to the knowledge of the fossil ostracodes of the Southern Ocean.

STUDY AREA
The Kerguelen Plateau is located in the Indian Ocean between 45 • S and 64 • S, north of the Antarctic Convergence.It lies in water depths between 1500 m and 2000 m, and about 2-3 km above the adjacent ocean basins Australian-Antarctica in the east, and African-Antarctic in the west (Fig. 1).Across the Kerguelen Plateau and along a latitudinal transect, six sites have been drilled at Ocean Drilling Program (ODP) Leg 199.Two of these (sites 738 and 744) were drilled in the southern part close to east Antarctica for documenting climatic changes imprinted in the sedimentary record.

MATERIALS AND METHODS
This study is based on the observation of 34 core samples of 10 cm 3 taken from Paleogene and Neogene sections of site 744A.The samples were disaggregated with water and diluted 100 vol.H 2 O 2 for a day, washed and wet sieved through a 63μm screen and, then, dried over a hot plate.Two samples from Late Eocene (119-744A-19H-5W-5 and 119-744A-19H-2W-5) have yielded more than 25 carapaces each, in contrast with less than five in some samples from Miocene and Pliocene section (Fig. 2).
The specimens figured in this article are housed at Museum of Paleontology of Universidade do Vale do Rio dos Sinos, under the curatorial numbers 7105 to 7135.In the taxonomy section, the following abbreviations are used: V (valve), LV (left valve), RV (right valve), h (height), l (length) and mbsf (meters bellow sea floor).Origin: 119-744A-19H-5W-5 (163.150mbsf).Age: Late Eocene.Material: one V.
Discussion: Maddocks (1969), in the revision on Bairdiidae, states that this is a widespread deep sea species with some degree of variability in the length and position of the posterior caudate extension, which could even correspond to more than one species or subspecies.The present specimen has both the hinge and duplicature poorly developed, being characterized as a juvenile.
The outline, shape, ornamentation and the presence of normal pore canals in the tubercles, clearly seen in Figure 11, plate 3 of Guernet (1993), led us to identify the species Henryhowella melobesioides (Brady) recorded by him as Anebocythereis hostizea (Hornibrook).

RESULTS
In this study, 28 species belonging to 14 genera and six families were identified.Krithe is the most diversified genus (eight spp.), followed by Cytherella and Dutoitella (three spp.for each).The ostracode incidence decreases from the bottom to the top of the section, being the peak of abundance and richness between the Late Eocene and the Early Oligocene.From the sample 119-744A-16H-4W5 of Early Oligocene age and younger ones there is a significant reduction in the richness and abundance.In most of these samples, the richness oscillates between one and two species, and the total abundance of this section is only 36 specimens (Fig. 2).The Early Oligocene threshold also depicts a faunal turnover, where 16 species only occur before this age, and six after it.Anebocythereis hostizea (Hornibrook) is the most abundant species and, with Bradleya johnsoni Benson, Legitimocythere presequenta (Benson), Krithe sp. 4, Krithe sp. 5, Henryhowella asperrima and Henryhowella sp., constitute the only species occurring both before and after the threshold.Some juvenile specimens of Krithe which were found in the majority of the studied samples, were not identified in the eight groups here presented, and their occurrences were not included in Figure 2.
The assemblages studied at this site present some similarity with the other faunal record of DSDP/ODP sites, in particular with the site 214, from Indian Ocean, studied by Guernet (1985).Three species are common to these two regions: Cytherella sp. 1, Cytherella sp. 2, and Dutoitella suhmi (Brady).Krithe sp. 2 is possibly cospecific with Krithe sp. 1 of Guernet (op. cit.,p. 287,pl. 1,fig. 16) but, due to the complex morphology of this genus, it is hard to sustain this assumption based only on Guernet's SEM pictures.Some slight variation in size was noticed in the species Bradleya johnsoni Benson, Agrenocythere hazelae (Bold) and Legitimocythere presequenta (Benson), and this is probably related to environmental conditions.In the Cenozoic, several climatic changes driven by orbital oscillations and their influences in the carbon cycle and glaciations have been recorded, which correspond to the transition from the Cretaceous greenhouse to the Cenozoic icehouse (Barker andThomas 2004, Zachos et al. 2001a).The Oligocene experienced a long glacial interval, except close to the Oligocene/Miocene boundary.Considering both geochemichal and orbital data, Zachos et al. (2001b) divided the Oligocene into four phases; the interval corresponding to the second and third ones (31 to 27 Ma) shows more positive 18 O signals, a factor that could explain at least in part the faunal threshold seen in the site 744A.
Positive peaks of 18 O in sea water are caused either by ice formation or cooling.Both have had different weight during Cenozoic events, and to find out which one was the most influent is not always straightforward (Lear et al. 2000).Considering that the ostracode faunal composition results from historic and oceanographic events, the cooling of the water and circulation changes in periods marked by 18 O peaks may influence both the evolution and migration of taxa prompting faunal turnovers.
Similar faunal trends have been found in the ostracodes from other ODP sites.Majoran and Dingle's (2002) study at the site 689 (Antarctica) recorded high values of richness and abundance in the Eocene-Oligocene interval, which they attributed to either taphonomic or hydrologic processes that resulted from the progressive cooling of Antarctica during that time.Guernet and Galbrun (1992) recorded at site 762 a high diversity and abundance of ostracodes from the Eocene to the Lower Miocene, and a sharp reduction from the Upper Miocene and younger ages.They did not propose any plausible explanation for this trend, but supposed that it could be a result of fluctuations of sedimentation rate linked to variations in the surface productivity.
The reduction in abundance seen in the upper portion of the studied section might be explained either by a preservational bias or a faunal impoverishment.Diester-Hass (1996) noticed a strong covariance between carbonate preservation and productivity in the Eocene-Oligocene interval in the Kerguelen Plateau: the increase in productivity was normally linked to an increase in carbonate dissolution, except when the region was under the influence of a warm, carbonate saturated water mass (WSDW-warm saline deep water).Hence, the carbonate preservation is strongly marked by the remodelation of oceanic circulation and productivity, and might have strongly influenced the fossil record in the upper portion of the section here studied.The presence of specimens (mainly Krithe) with a variable degree of dissolution sustains this hypothesis.A similar cause could explain the scarcity of fossils in the younger samples studied (Late Oligocene onwards), in as much as no other process would easily explain the fossil record pattern.

INTRASPECIFIC VARIATION IN DEEP SEA OSTRACODES
The discussion on the intraspecific variation in ostracodes pervades the fields of ecology and systematics.In their discussion on deep sea ostracodes diversity, Jellinek and Swanson (2003) state that a precise taxonomic approach would not be achieved based exclusively on the carapace morphology, at least in some ostracod groups (Trachyleberididae, for instance).The refinement of the taxonomic knowledge on deep sea ostracodes is the basis for their paleoceanographical use, and recent studies show that much has to be done in this field.Schornikov (2005), for instance, concluded that at least five species were lumped under the name Pedicythere polita Colalongo and Pasini around the world, making them so called composite species.
However, species such as Krithe dolichodeira Bold, Legitimocythere presequenta (Benson) and Agrenocythere hazelae (Bold), actually have near global distributions in the deep ocean.In these species, slight morphological variations are present especially on size and ornamentation, as can be seen even in this study.Evidences from the previously discussed studies sustain that intraspecific variation could also be a common phenomenon in deep sea faunas, which is resulted not only from clinal variation, but also induced by change in temperature, dissolved oxygen and salinity.
Considering that climatic changes exert influence on deep sea ostracodes even for a short geological duration (Cronin et al. 1999, Yasuhara et al. 2008), it would be plausible to find climatically driven ecophenotypic variants of a species in these environments.This can be achieved only through an accurate taxonomic knowledge and the understanding of the intraspecific variation processes, reinforcing the use of ostracode diversity as a proxy for hydrological changes.

Fig. 2 -
Fig. 2 -Occurrence and abundance of species in the samples studied.The shadowed column refer to the faunal threshold.
Mazzini (2005)na Guernet and Moullade 1994Type species Cytheropteron mucronalatum Brady 1880 instance, is hardly achieved in many studies, due to either the poorly precise descriptions of the type material or the inadequacy of their original illustrations.The widespread use of the taxonomic terms aff., cf. or gr. is a testimony of this problem.The present material is considered cospecific to the topotypic material figured byMazzini (2005).The spinosity, reticulation and a ventro-lateral spinose ridge in the present species would allow its inclusion in Legitimocythere Coles and Whatley.However, based on the discussion presented by Jellinek and Swanson op.cit.about the age of the genotype elected for this genus, we prefer not to adopt it for the present species.According to the age and geographic distribution, the genus Taracythere Ayress seems to be a more suitable option.
Discussion: The taxonomy of the genus Henryhowella has been the subject of intense discussion.The accurate identification of the species H. asperrima and H. evax, An Acad Bras Cienc (2010) 82 (3) for An Acad Bras Cienc (2010) 82 (3)