Abstract

The end of the Paleocene Epoch was marked by an abrupt, worldwide extinction of deep-sea benthic organisms. At about 55 Ma, between 30 and 50% of the benthic foraminifers suddenly became extinct, in association with comparable ostracode extinctions. Extinctions of planktonic taxa were insignificant. This extinction event is considered the largest of the past 90 million years (m.y.) in the deep-sea. Although of major proportions, this biotic crisis was almost unknown before the past decade because it had little effect on shallow marine invertebrates and ocean plankton. This was a "bottom-up" extinction, compared with the "top-down" extinction that marked the Cretaceous/Tertiary boundary.

High-resolution stratigraphic studies in deep-sea sediments indicate that the extinction occurred in less than 3000 yr. Foraminiferal oxygen and carbon stable isotope changes, in combination with a distinct change in benthic fauna, indicate an abrupt but temporary warming and oxygen depletion of deep waters related to a fundamental change in oceanic circulation. The available data point to an ocean temporarily dominated by warm saline deep water whose source was probably in the middle latitudes.

Although the terminal Paleocene environmental changes were relatively brief, this transient global warming event affected both ocean and terrestrial spheres simultaneously and had a great influence on the course of global biotic evolution for the remainder of the Cenozoic.

Introduction

Much interest exists in the character and causes of major biotic extinctions in the geologic past. The stratigraphic record demonstrates that the biosphere, or parts of it, have experienced major disruptions in the geologic past, some involving mass extinctions. Mass extinctions involve a sudden, and short-lived, increase in extinction rates well above normal background levels, and can affect a great variety of biotic groups (Flessa, 1990). Numerous theories have been proposed to explain mass extinctions (for reviews, see Stanley, 1984, 1987; Hallam, 1989). All involve a response by the biosphere to radical changes in the environment on regional or global scales. However, a persistent problem requiring resolution is how to explain contemporaneous extinctions over broad areas of the Earth's surface and over a wide range of habitats. Why did certain groups of taxa, particularly those that were abundant over large areas of the Earth, suddenly cease to exist? What factors control the timing of mass extinctions and the rate of biotic turnover? Theories advanced to explain mass extinction are of two general categories. The first, and perhaps most popular, has invoked extraterrestrial causes, particularly massive global environmental change resulting from bolide impacts on Earth (Alvarez et al., 1980). The second involves intrinsic changes exclusively within the Earth's environment (Stanley, 1984). General popularity for an extraterrestrial cause of mass extinction stems in part from the fact that it seems to provide a mechanism for sufficiently large and rapid changes in the global environment. In contrast, it appears more difficult to explain how the Earth's environment might have changed intrinsically on a magnitude necessary to cause mass extinction. Have intrinsic changes in the global environment ever been large and rapid enough to cause biotic crises of this scale? Stanley (1984) argued that most mass extinctions have resulted from climatic change, particularly cooling. Debate continues, however, about the relative merits of extrinsic and intrinsic causes.

The purpose of this essay is to briefly summarize evidence for the possible cause of a mass extinction of deep-sea biota near the end of the Paleocene about 55 million years ago. Existing data suggest that the extinction event resulted from large, rapid changes within the Earth's environmental system without extraterrestrial forcing.

Evidence for mass extinction comes exclusively from the stratigraphic record, and the quality of the stratigraphic data, including their resolution, is usually the key to better understanding of causes. Critical information includes the rate of extinction; which sectors of the biosphere were involved; description of the taxa that did or did not become extinct; the sequence of extinctions in taxa; and relationships of the extinction event to a wide variety of paleoenvironmental proxies. Relatively few high-quality sediment sequences are available that are sufficiently fossiliferous and were deposited continuously at high enough rates of sedimentation to provide the required resolution. Numerous stratigraphic sections contain hiatuses of various duration contemporaneous with major extinctions events. Such disruption in the stratigraphic record probably resulted from sediment erosion related to changes in oceanic circulation and/or sea-level change at times of major global environmental change.

During the past two decades, the Deep-Sea Drilling Project (DSDP) and its successor, the Ocean Drilling Program (ODP), have provided fossil-rich ocean sediment sequences covering vast, otherwise inaccessible, areas of the Earth's surface, including the high latitudes. The problem of mass extinction has thus been assisted by the availability of a broader array of sequences that record such events and of critical new information about changes in the deep-sea environments and biota. The ocean ecosystem deeper than the continental shelf is vast, forming more than 90% by volume of the Earth's habitable environments (Childress, 1983). Questions about mass extinctions require information about any response or role played by the deep-sea habitat.

The Paleocene-Eocene transition has long been differentiated by stratigraphers based on significant biotic changes at the end of the Paleocene. However, knowledge of stratigraphic relationships between marine and terrestrial records has been hampered by poor chronostratigraphic control. The primary problem has been the discontinuous nature and poor biostratigraphic control of the classic Late Paleocene-Early Eocene European stratotype sections that form the foundation of global stratigraphic correlations. Terrestrial to marine correlations have been improved with the application of carbon isotope stratigraphy.

During the Paleocene and Early Eocene there were large, systematic patterns of δ¹³C variability in the ocean that have been correlated globally (Shackleton and Hall, 1984; Stott etal., 1990; Pak and Miller, 1992; Stott, 1992; Zachos et al., 1993a,b). These carbon isotope (δ¹³C) variations reflect changes in the ¹²C/¹³C ∑CO₂ in the ocean. Because the ocean and atmospheric reservoirs of CO₂ tend to maintain approximate isotopic equilibrium, variations in the ocean's δ¹³C composition will be transmitted to the terrestrial reservoirs of carbon via the atmosphere. Soil carbonates, freshwater fossils, and terrestrial biomass, therefore, also exhibit the large carbon isotopic variations of the marine fossil record. The absolute isotopic values differ among these various terrestrial and marine carbon reservoirs due to systematic differences in the fractionation of ¹²C and ¹³C. However, these fractionation patterns are known and can be used to predict isotopic stratigraphies in terrestrial sections. Hence, the large-scale patterns of δ¹³C variability recorded in the marine sections across the Paleocene-Eocene boundary are now being discovered in terrestrial sections (Koch et al., 1992; Sinha and Stott, 1994).

With this new global stratigraphy it has become apparent that accelerated evolution in terrestrial mammals during the Late Paleocene coincided with the extinction and environmental changes recorded in the deep sea (Rea et al., 1990; Koch et al., 1992). Although this chapter is concerned only with the marine record of extinction, the terrestrial record cannot be viewed independently. The abrupt climatic warming at the end of the Paleocene may have stimulated the evolution of major new mammalian groups such as the artiodactyls, perissodactyls, and the primates (Koch et al., 1992). Within the next few years it should be possible to integrate the terrestrial and marine records of faunal change at sufficient stratigraphic resolution to provide perhaps the best example of biotic change associated with abrupt environmental change.

Originally, observations on biotic change in the Paleocene-Eocene transition were limited to shallow marine invertebrate and terrestrial fossils. It has been generally assumed that, as with other epoch boundaries, changes in the global environment produced the biospheric response. Until recently the character of these changes remained largely unknown. Abrupt global warming and associated environmental changes now known for the end of the Paleocene are clearly implicated as a cause of this biotic crisis.

Terminal Paleocene Mass Extinction in the Deep Sea

The oceanic deep-sea sediment record of the Paleocene-Eocene transition is clearly marked by major deep-sea benthic foraminiferal extinctions, perhaps the largest of the past 90 m.y. (Thomas, 1990). This event profoundly affected oceanic benthic communities deeper than the continental shelf (>100 m; neritic zone) resulting in a 35 to 50% species reduction in benthic foraminifera (Thomas, 1990). Deep-sea benthic foraminiferal assemblages radically changed as a result of this event. Late Paleocene assemblages before the extinction are highly diverse and contain genera with long stratigraphic ranges through the Late Cretaceous and Paleocene. Indeed, as pointed out by Thomas (1990, 1992), deep-sea benthic foraminiferal assemblages were little affected during the massive extinctions at the Cretaceous-Tertiary boundary (K/T), about 8 m.y. earlier. Instead many cosmopolitan benthic foraminiferal taxa typical of late Mesozoic and Paleocene assemblages, including all species within certain genera, were eliminated at the end of the Paleocene.

Until recently, the Paleocene-Eocene boundary was not generally recognized as a time of major biotic crisis, because generic-level extinction rates were low (Raup and Sepkoski, 1984). These patterns of extinction, however, are from the shallow marine and terrestrial spheres, not the deep-sea (Thomas, 1992).

The mass extinction that predates the Paleocene-Eocene boundary is located near the middle of a long reversed polarity interval, identified as Magnetochron 24 R (Miller et al., 1987; Stott and Kennett, 1990), within the latest Paleocene. The position of the Paleocene-Eocene boundary appears to be about 40,000 yr younger, but also within Chron 24 R (Kennett and Stott, 1991). Correlation of these respective levels between high and low latitude sequences is still inadequate to establish the precise age and prove synchroneity (for detailed discussion see Thomas, 1990; Kennett and Stott, 1991; Miller et al., 1992). Ages applied to the extinction differ among different sequences, ranging between -57.33 and 58 m.y. ago (Ma) (Miller et al., 1992). However, for several reasons discussed below, we believe that the mass extinction was synchronous and that the different age assignments have resulted from the current inadequacy of biostratigraphic correlations using planktonic microfossils across different latitudes and oceans. This appears, in part, to be the result of diachronism in planktonic microfossil datums and/or insufficient resolution in the biomagnetostratigraphic schemes employed. The age that we previously adopted for the mass extinction was 57.33 Ma (Kennett and Stott, 1991). However, the time scale of the magnetostratigraphic sequence in the vicinity of the extinction is now considered to be 2 m.y. younger (Cande and Kent, 1992).

The mass extinction is marked by four types of change in benthic foraminiferal assemblages: (1) the extinction of many taxa, (2) diversity decrease, (3) changes in the relative abundance of taxa, and (4) a general decrease in test size of taxa within the assemblages. As a result, the benthic faunas were distinctly different before and after the mass extinction.

The extinction removed species within the genera Stensioina, Neoflabellina, Bolivinoides, Pyramidinia, Pullenia, Aragonia, Tritaxia, Gyroidinoides, Neoeponides, Quadratobulimina, Stilostomella, Dorothia, and many other forms making up what has been termed the S. beccariiformis assemblage. Many of the extinct forms were morphologically distinctive, and their loss from the deep-sea record adds to the conspicuousness of this datum level. Epifaunal forms were particularly devastated (Thomas, 1990).

The extinction was first recognized in upper bathyal marine sequences from Trinidad (Beckmann, 1960); the Richenhall and Salzburg Basins, Austria (von Hildebrandt, 1962); and northern Italy (Braga et al., 1975). With the recognition of the extinction in the deep-sea by Tjalsma and Lohmann (1983), as well as Schnitker (1979), Kaiho (1988), Berggren and Miller (1989), Nomura (1991), and Thomas (1990, 1992), its global character was clearly established. It is also evident that a wide range of paleodepths were affected, ranging from bathyal to abyssal depths (Miller et al., 1987).

Despite the global distribution of this extinction horizon, stratigraphic resolution was too low to determine whether the event was geologically instantaneous throughout the oceans or diachronous. More recent biostratigraphic investigations in conjunction with stable isotope stratigraphy (Kennett and Stott, 1991; Thomas, 1992) have strengthened the concept that the extinction was synchronous throughout the oceans. Nevertheless, this still requires confirmation with high-resolution studies at numerous sequences that can be accurately correlated by using carbon isotopic, paleomagnetic, and biostratigraphic data (Sinha and Stott, 1994).

Immediately following the extinction, the benthic foraminiferal assemblages were dominated by small, thin-walled specimens (Thomas, 1990). Benthic taxa that survived the extinction included Nuttallides truempyi, which became a dominant component in the Eocene, as well as Bulimina semicostata and other taxa making up what has been termed the Nuttallides truempyi assemblage. This new, relatively low-diversity assemblage includes about six forms that dominated Early to Middle Eocene benthic foraminiferal assemblages (Tjalsma and Lohmann, 1983; Miller et al., 1987). Faunal assemblages following the extinction are less cosmopolitan (Thomas, 1990).

Although the extinction event was abrupt, there is some evidence that the S. beccariformis assemblage became progressively restricted to shallower depths during the Paleocene (Tjalsma and Lohmann, 1983; Miller et al., 1987). Also, the relative abundances of certain forms in this assemblage decreased during the Late Paleocene and were replaced by forms more typical of the Nuttallides truempyi assemblage of latest Paleocene to Early Eocene age (Miller et al., 1987). These changes culminated at the mass extinction and suggest that some form of biological threshold was surpassed.

For several million years following the extinction there occurred a radiation of benthic foraminiferal taxa. These probably filled vacancies left by the latest Paleocene extinctions (Tjalsma and Lohmann, 1983; Miller et al., 1987). The postextinction assemblages included long-ranging forms such as Pullenia bulloides and Globocassidulina subglobosa (Thomas, 1990). The radiation caused a diversity increase that peaked during the early Middle Eocene. Nevertheless, the high diversity values of the Late Cretaceous and Early Paleocene were never attained again (Thomas, 1990).

In contrast to the benthic assemblages, oceanic planktonic microfossil assemblages underwent no mass extinction at the end of the Paleocene, but did exhibit distinct change in the species composition in the Antarctic. A general increase in diversity marks the Late Paleocene high- to middle-latitude assemblages of planktonic foraminifera, calcareous nannofossils, and dinoflagellates (Premoli-Silva and Boersma, 1984; Oberhänsli and Hsü, 1986; Stott and Kennett, 1990; Pospichal and Wise, 1990). This increase in diversity stemmed, in part, from the incursion of lower-latitude groups into the Southern Ocean. The diversity increase at the end of the Early Paleocene was superimposed on a longer-term increase that began during the Paleocene, following the K/T boundary extinctions (Corfield, 1987). The plankton diversity increase may have been caused by the increased surface water temperatures at high- to middle-latitude regions. This increase in surface water temperatures was particularly pronounced in the Antarctic during the latest Paleocene, as reflected by the relatively brief appearance of the subtropical-tropical morozovellid group and a peak in discoaster abundance (Pospichal and Wise, 1990; Stott and Kennett, 1990). The emigration of these warm-loving planktonic microfossils to the Antarctic was particularly pronounced during the mass extinction. In one Antarctic site 32% of the planktonic foraminiferal species appeared for the first time in the latest Paleocene, 27% underwent major abundance changes, and only 13% were eliminated from the assemblages (Lu and Keller, 1993). Most new entries were surface dwellers. Of those that were eliminated, most were deeper dwellers such as the subbotinids (Lu and Keller, 1993). Coeval low latitude planktonic assemblages underwent little change (Miller et al., 1987; Miller, 1991) presumably because of the relatively stable sea surface temperatures (Stott, 1992).

Association Between Mass Extinction and Oceanic Warming

In earlier work (Kennett and Stott, 1990; Stott et al., 1990) we discovered a dramatic negative oxygen and carbon isotopic excursion of brief duration (Figure 5.1) that coincided closely with the terminal Paleocene benthic foraminiferal extinction event in an Antarctic Paleogene sequence (ODP Site 690B) (Thomas 1989, 1990). This discovery stimulated a high-resolution study of the extinction event (Kennett and Stott, 1991). Results from that study demonstrated the intimate temporal relationship between the mass extinction and a large oxygen and carbon isotope excursion in both benthic and planktonic foraminifera. Planktonic values of δ¹⁸O abruptly decreased by 1.0 to 1.5%o, and by about 2%o in the benthics; values of δ¹³C also decreased by 4%o in surface-dwelling planktonic foraminifera, and -2%o in the deeper-dwelling planktonic and benthic forms. The planktonic foraminifer Acarinina praepentacamerata records the lowest oxygen and highest carbon isotope values within the excursion, consistent with an inferred near-surface habitat (Stott et al., 1990; Kennett and Stott, 1991). The species of Subbotina record higher δ¹⁸O and lower δ¹³C values, indicating a deep water planktonic habitat and/or a preference for cooler months of the year (Stott et al., 1990; Kennett and Stott, 1991). The highest δ¹⁸O and lowest δ¹³C values are exhibited by the benthic foraminifer Nuttalides truempyi, reflecting its habitat in relatively nutrient-rich high latitude deep water.

Figure 5.1

Composite oxygen isotopic record of planktonic foraminifers of ODP Sites 689 and 690 in the Antarctic Ocean (from Stott et al., 1990). Note the abrupt negative excursion near the Paleocene-Eocene boundary. Time scale follows that of Berrgren et al. (more...)

During the latest Paleocene interval immediately preceding the extinction (before ~55.33 Ma), surface water temperatures estimated from oxygen isotopic values were -13 to 14°C (Figure 5.2). Deep water temperatures were -10°C at about 2100 m. Thus, before the extinction, little temperature difference existed between surface and deep water in the Antarctic.

Figure 5.2

Changes in oxygen and carbon isotopic composition of planktonic foraminifera (Acarinina praepentacamerata and Subbotina) and a benthic foraminifer (Nuttalides truempyi) in the latest Paleocene (55.0 to 55.6 Ma) in relation to mass extinction in benthic (more...)

The excursion began abruptly at 55.33 Ma, with conspicuous decreases in δ¹⁸O and δ¹³C values, followed by a return to values only slightly lower than those before the excursion (Figure 5.2). The δ¹⁸O and δ¹³C changes are reflected in all three foraminiferal taxa, although at different amplitudes (Figure 5.2). The significance of these differences is discussed by Kennett and Stott (1991). We limit our discussion here to the trends of critical paleoenvironmental importance. The largest δ¹⁸O shift (2.0%o) at the beginning of the excursion is recorded by the benthic foraminifera, an intermediate shift (1.5%o) by deeper-dwelling planktonics, and the smallest shift (1.0%o) by shallow-dwelling planktonic forms (Figure 5.2). The initial oxygen isotopic shift exhibited by the surface-dwelling form, which coincided with the mass extinction, possibly reflects an increase in surface water temperatures from 14 to 18°C. This was followed by an additional δ¹⁸O decrease of -1.0%o, indicating a further possible temperature increase in surface waters to 22°C. The brief interval represented by the excursion was likely the warmest of the entire Cenozoic although, as discussed later, some fraction of the decrease in δ¹⁸O values may have resulted from reduction in surface water salinity.

Of great significance, however, is the observation that the largest δ¹⁸O change was recorded by the benthic, rather than the planktonic, foraminifera (Figure 5.2). Thus, deep waters warmed more than surface waters. For a brief interval, beginning at -55.31 Ma, deep waters had warmed to such a degree that the temperature gradient between deep and surface waters was virtually eliminated at this location in the Antarctic region. The extinctions occurred at the beginning of the temperature excursion.

The initial, rapid temperature rise encompassed -3000 yr and was followed by a more gradual decrease in ocean temperatures at all water depths. At the end of the excursion, the water column in this Antarctic region was only slightly warmer than it had been immediately before the excursion less than 100,000 yr earlier.

The magnitude of carbon isotopic change between the planktonic and benthic foraminfera was different from that of oxygen isotopes. Whereas the benthic (bottom dwellers) recorded the largest δ¹⁸O change, it was the plankton that recorded the largest δ¹³C change (4%o). The 4%o shift in δ¹³C is the largest so far known for the Cenozoic Period. The magnitude of the shift clearly underscores the significance of this event. During the brief interval at -55.32 Ma when the vertical δ¹⁸O gradient was eliminated, the previously large surface to deep water δ¹³C gradient was also almost completely eliminated. The cause of the δ¹³C change remains enigmatic. However, Stott (1992) presented evidence that the δ¹³C of marine organic matter became more positive at the time of the excursion. If this was a global phenomenon, it would suggest that the negative δ¹³C excursion recorded in foraminiferal calcite resulted from a redistribution of δ¹²C between photosynthetic organic matter and the inorganic pool of carbon in the Late Paleocene oceans.

It is clear from this Antarctic record that the mass extinction coincided with the beginning of the sharp, negative shifts in δ¹⁸O and δ¹³C. However, to determine whether all the extinctions occurred simultaneously and in conjunction with the initiation of the δ^δO and δ¹³C change, Kennett and Stott (1991) increased the sample resolution to only 1-cm intervals (-800 yr) across the extinction interval in Site 690B (Figure 5.3). This was possible because the interval was not bioturbated (Kennett and Stott, 1991). Before the extinction, benthic foraminiferal assemblages (>150-µm fraction) were diverse, averaging about 60 species, or even more (Thomas, 1990). Assemblages included an abundance of forms interpreted to be of both infaunal and epifaunal habit (Corliss and Chen, 1988; Thomas, 1990). This included a high diversity of trochospiral and other coiled forms.

Figure 5.3

Changes in oxygen and carbon isotope composition of the planktonic foraminifers A. praepentacamerata and Subbotina patagonica, and in simple diversity of benthic foraminiferal assemblages at high stratigraphic resolution over the latest Paleocene mass (more...)

The extinction in Site 690B (Figure 5.3) involved a rapid drop in benthic foraminiferal diversity (>150 µm) from ~60 to 17 species, representing a diversity reduction of 72% within 3000 yr (4 cm). Many distinct taxa such as Stensioina beccariiformis and Neoflabellina disappeared early, during an interval of less than 1500 yr (Figure 5.3). Most trochospiral forms such as Stensioina had disappeared by the midpoint of the oxygen isotopic shift. The survival of a higher proportion of infaunal forms indicates some advantage over the epifaunal forms that lived at or close to the sediment-water interface. Nevertheless, the infaunal environment was not entirely unaffected since many of the taxa inferred to have been living there also disappeared. The abundance of benthic foraminifera (>150-µm fraction) was also severely reduced, although small individuals (<150 µm) remained abundant throughout. The ostracoda also exhibit a drastic decrease in diversity and abundance. The benthic foraminiferal assemblage was strongly depleted of coiled forms for several thousand years following the extinction. This left assemblages (>150 µm) temporarily dominated by relatively small, thin-walled, uniserial, triserial, and other forms more typical of an infaunal habitat (Corliss and Chen, 1988; Thomas, 1990). The relative increase in abundances of small benthic foraminiferal specimens, associated with a decrease in diversity, suggests conditions low in oxygen and higher in nutrients (Bernard, 1986; Thomas, 1990).

Following a brief interval of extremely low diversity and abundance in the fraction greater than 150 µm (Figure 5.3), diversity increased again to about 30 species on average—a diversity of about half that before the extinction. This increase seems to have resulted mainly from the reappearance of forms that had been temporarily excluded from the benthic foraminiferal assemblage. Nevertheless, about 35% of the Late Paleocene species at Site 690B became completely extinct.

Cause of Mass Extinction In Deep Sea

It is clear that the mass extinction was restricted to the deep-sea biota deeper than the continental shelf or the thermocline. The lack of major extinctions in the oceanic planktonic and shallow water benthic communities strongly suggests that the extinctions were not caused by an extraterrestrial impact with the Earth, as has been implicated for the terminal Cretaceous extinctions (Alvarez et al., 1980). An intrinsic oceanic cause is considered more likely (Kennett and Stott, 1991). The process that caused the mass extinction must have had the capacity to strongly affect the vast volume of the deep ocean in an interval of less than 3000 yr. Indeed, the extinctions may well have taken place at the rate of replacement time of the oceans, currently about 1000 yr, although this was possibly slightly slower during the early Paleogene.

The superposition of the abrupt, negative δ¹³C and δ¹⁸O shifts upon similar, more gradual trends during the Late Paleocene (-60 Ma) to Early Eocene (-55 Ma) (Figure 5.1) suggests the involvement of a climatic threshold event similar to the oxygen isotopic shift near the Eocene-Oligocene boundary, although in an opposite sense (Kennett and Shackleton, 1976). The speed and magnitude of the associated temperature increase imply global warming with strong positive feedback mechanisms, not just warming restricted to the oceans. Indeed, isotopic fluctuations in the marine carbonate record are closely tracked by the terrestrial records provided by paleosol carbonates and mammalian tooth enamel (Koch et al., 1992).

Three main hypotheses have been proposed to account for this mass extinction. These are (1) the rapid warming of deep waters (Miller et al., 1987); (2) an oxygen deficiency in deep waters resulting from the sudden warming and change in deep-sea circulation (Kennett and Barker, 1990; Kennett and Stott, 1990a; Thomas, 1990, 1992; Katz and Miller, 1991); and (3) a sharp drop in surface ocean biological productivity that reduced the supply of organic matter, the food source of deep-sea benthic organisms, initiating a cascading trend of food chain collapse (Shackleton et al., 1985; Shackleton, 1986; Rea et al., 1990; Stott, 1992). If this had occurred, significant changes in oceanic plankton would be expected as well. There is no suggestion that this happened in the carbonate groups. It is also likely that reductions in abundance would have occurred in the infaunal deep-sea benthic foraminiferal assemblages. Indeed, Thomas (1990) showed that during the mass extinction, an increase occurred in the abundance of small infaunal species. Apparently these types of benthic foraminifera occur where there is availability of sedimentary organic carbon. Therefore, it is more likely that a greater abundance of these taxa resulted from an increase rather than a decrease in organic productivity, or a decrease in the oxygen content of deep waters, resulting in decreased oxidation of organic matter.

A general consensus exists that the mass extinction was caused by a rapid temperature increase of deep waters or the environmental effects associated with higher temperatures, including reduction in oxygen concentrations (Miller et al., 1987; Thomas, 1989, 1990, 1992; Kennett and Stott, 1990, 1991; Stott et al., 1990). Deep waters at the time of the excursion warmed to ~18°C. At such temperatures, deep waters would almost certainly have been depleted in oxygen even if oceanic primary production was lower and atmospheric oxygen levels were higher (Kennett and Stott, 1991). Widespread dysaerobism occurred in the deep ocean. However, there is no evidence that deep waters became completely anoxic, which would have caused an increase in accumulation of organic carbon during the excursion. Furthermore, infaunal benthic foraminifera remained abundant, which would not have been the case if there had been complete anoxia (Bernard, 1986). The dysaerobism was not as extreme as during the Cretaceous "oceanic anoxic events" (Schlanger and Jenkyns, 1976).

Cause of Oceanographic and Climate Change

The available data point to a mass extinction at the end of the Paleocene, resulting from physicochemical environmental changes in the deep-sea, especially rapid warming and a decrease in oxygen concentrations. What ocean process could have created such widespread and rapid change? Consensus has developed (Miller et al., 1987; Thomas, 1990, 1992; Kennett and Stott, 1990a, 1991; Katz and Miller, 1991) that the rapid deep-sea warming resulted from a rapid change to near dominance in the deep ocean, of warm saline deep waters (WSDW) produced in the middle-latitude regions (see also Mead et al., 1993). At the same time it is believed that there was a severe reduction in the production of deep waters produced at high latitudes. Such ocean circulation, drastically different from that of the modern ocean, was termed Proteus by Kennett and Stott (1990b; Figure 5.4). In the present ocean, most bottom waters are formed at high latitudes where cold temperatures, in combination with moderately high salinities, cause waters to become dense and sink (for summary see Broecker and Peng, 1982). These waters are relatively oxygen rich. At the same time, warm saline dense waters are formed in the modern Mediterranean and Red Seas as a result of high net evaporation. Because of low buoyancy fluxes, these waters sink only to thermocline depths. These waters do not represent large volumes in the modern deep ocean, although warm saline waters produced in the Mediterranean eventually become an important component of North Atlantic deep water (NADW) (Reid, 1979). Brass et al. (1982) suggested that this warm deep water (>10°C) of the Cretaceous and Early Paleogene reflected production of warm saline deep waters in middle-latitude areas.

FIGURE 5.4

General model for deep and intermediate water circulation proposed for the time of the terminal Paleocene isotopic excursion and mass extinction. This model has been termed Proteus by Kennett and Stott (1990b). This is compared with the general circulation (more...)

The rapid ocean warming associated with the isotope excursion and mass extinction must have occurred from the deep ocean upward. We believe that this was caused by the elimination of deep water formation at high latitudes and the incursion of warm water from middle latitudes.

Synchroneity (Figure 5.3) of the mass extinction, the negative δ¹⁸ O shift, and the negative δ¹³C shift in the deep-dwelling planktonic foraminifer Subbotina patagonica is apparent. Calculations of rates of sedimentation suggest that the mass extinction occurred in less than about 3000 yr, the oxygen isotopic shift in less than 4000 yr, and the carbon isotopic shift in less than 6000 yr. The mass extinction occurred at the beginning of the oceanographic and climatic changes that mark the excursion. Most of the large δ¹³C shift occurred slightly later in surface waters (Figure 5.3).

Why the Early Paleogene?

No such deep oceanic mass extinction and isotopic excursion has yet been discovered at any other time in the Cenozoic. This was an unusual, if not unique, event. Why did the event occur at about 55 Ma, early in the Cenozoic, rather than later? During the early Paleogene, different global geography and climate (Kennett, 1977; Haq, 1981; Hay, 1989) combined to make ocean circulation distinct from that of modern and, indeed, Neogene oceans (Kennett, 1977; Benson, 1979; Kennett and Stott, 1990a). Much evidence exists for relatively warm climates in the Antarctic region during the early Cenozoic (Kennett and Barker, 1990). Oxygen isotopic data suggest average Antarctic Ocean surface water temperatures of ~14°C during the Late Paleocene. Decreased meridional thermal gradients led to a decrease in global zonal wind intensity (Janecek and Rea, 1983; Hovan and Rea, 1992). Clay mineral assemblages in offshore sequences derived from the Antarctic continent were formed predominantly by chemical weathering under conditions of relative continental warmth and humidity (Robert and Kennett, 1992). Extensive coastal cool temperate rain forests dominated by Nothofagus indicate high continental rainfall and a lack of perglacial conditions at sea-level (Case, 1988). Ice-rafted sediments are absent, as is other evidence for continental cryosphere of any extent (Kennett and Barker, 1990), although montane glaciation seems probable. The Antarctic Ocean was dominated by calcareous planktonic microfossil assemblages of high diversity rather than siliceous forms (Kennett, 1977). Faunas were cool to warm temperate in character. Deep waters in the global ocean were warm, averaging 10 to 12°C compared with ~2°C in the modern ocean (Shackleton and Kennett, 1975; Stott et al., 1990). The Earth was clearly in a ''greenhouse" mode—a condition that appears to have been much exaggerated during the terminal Paleocene isotopic excursion. Relatively high precipitation in the Antarctic region at this time is inferred to have contributed to a large reduction in deep water production at high latitudes (Kennett and Stott, 1991).

At the same time, the extensive mid-latitude Tethys Seaway north of Africa was a likely location for the production of large volumes of warm saline deep waters (Kennett and Stott, 1990). Tectonic reconstructions (Dercourt et al., 1986) show that the Tethys Seaway in the early Cenozoic contained extensive shallow carbonate platforms with dolomites and evaporitic sediments. During the excursion, these various factors combined to form, through positive feedback responses, an extreme case of the Proteus Ocean, an ocean dominated by middle-latitude deep water production (Kennett and Stott, 1990, 1991). Climate model studies (Barron, 1987; Covey and Barron, 1988) suggest that large-scale meridional heat transport became more effective via the deep oceans relative to the atmosphere (Hovan and Rea, 1992).

The forcing mechanism of ocean warming and associated faunal turnover at the end of the Paleocene is not yet known. Rea et al. (1990) have suggested that the abruptness of environmental changes and the associated mass extinction were possibly triggered by rapid input of CO₂ into the atmosphere from volcanism and/or hydrothermal activity that was extensive over the Paleocene-Eocene transition. The warming of the oceans would have been the most obvious effect of enhanced greenhouse forcing resulting from this. Whether or not this would have been sufficiently rapid and large to explain the rapid rise in temperatures associated with the extinctions at 55 Ma remains to be tested. The triggering mechanism for the rapid climate change at the end of the Paleocene remains unknown.

In one attempt to test whether CO₂ might be implicated in the oceanic warming, Stott (1992) presented evidence that the Paleocene ocean-atmosphere system was indeed associated with higher levels of CO₂ compared to the present time. However, on the basis of the same data it appears that the extinction interval was actually associated with lower oceanic CO₂, not higher. How could an abrupt warming at the end of the Paleocene be associated with lower oceanic CO₂? The answer may lie in the way the ocean and atmosphere cycle CO₂. The problem is that the solubility of CO₂ in seawater decreases with increasing temperature. The exchange of CO₂ between the ocean and atmosphere was further complicated by changes in atmospheric circulation occurring at that time, which would have affected turbulence of the mixed layer ocean. This, together with changes in the biological pump (photosynthesis), constitute factors that are not yet well constrained for the Paleocene-Eocene. However, it is evident that with the high sea surface temperatures at the end of the Paleocene, particularly in regions of normal deep water advection (e.g., high latitudes), the oceans were probably less efficient in taking up CO₂ from the atmosphere.

Implications and Summary

A conceptual model of the possible chain of environmental events related to the terminal Paleocene mass extinction in the deep-sea is shown in Figure 5.5. The extinction occurred in less than ~3000 yr and was associated with global deep-sea warming of similar rapidity (Figure 5.3). Both benthic foraminifera and ostracoda were se verely affected, although a synchronous decrease in bioturbation during the extinction at one location suggests that nonskeletal benthic organisms were also severely affected. Apparently the organisms that became extinct were unable to cope with the rapidity and magnitude of deep-sea warming and associated depletion in oxygen levels. Brief elimination of the vertical δ¹⁸O and δ¹³C gradients during the extinction event indicates vertical ocean mixing and homogenization of nutrient distributions over a large depth range at high latitudes (Figure 5.5). It seems that this was related to temporary instability of the water column and even ocean turnover at high latitudes. This did not seem to be the case in the tropics.

Figure 5.5

Conceptual model of possible chain of environmental events at the time of the mass deep-sea biotic extinction near the end of the Paleocene (55.33 Ma).

The character of the oxygen isotopic changes associated with the mass extinction indicates a temporary switch to almost total dominance of warm saline deep water and associated interruption in the production of deep waters at high latitudes (Figure 5.4). Surface waters warmed significantly at high latitudes but little in the tropics. The major δ¹³C shift reflects large, rapid changes in the distribution of CO ₂ and nutrients in the ocean during the extinction. The magnitude of the δ¹³C shift in Antarctic surface waters suggests major changes in nutrients and partial pressure of CO₂ in surface waters. This may imply an associated increase in atmospheric CO₂ and resulting global greenhouse warming (Figure 5.5).

Rapid global warming would have led to increased transfer of heat from low to high latitudes, particularly via the oceans. Increased sea-surface temperatures would have caused an increase in atmospheric saturation vapor pressure. Latent heat transfer would have increased from the tropics to the poles, causing a significant increase in rainfall in the Antarctic region (Figure 5.5). Clay mineralogical evidence from the Antarctic suggests the occurrence of such an increase in rainfall. Increased rainfall in the Antarctic region would have reduced ocean surface water salinity. This in turn, would have contributed to a decrease in the production of Antarctic deep waters to the world ocean, thus reinforcing the dominance of warm saline deep waters (Figure 5.5).

A number of broad implications are suggested as a result of these discoveries. The extinction resulted from changes entirely within the Earth's environmental system, in the apparent absence of any extraterrestrial influences. Mass extinctions can be produced by large changes in the ocean. Such changes can occur very quickly under certain conditions. Mass extinction can be restricted to certain parts of the Earth's ecosystem and be effectively decoupled from other parts of the biosphere. In the case of the terminal Paleocene mass extinction, there was almost total decoupling between the deep and shallow marine ecosystems. Despite significant rapid warming in shallow and deep environments the mass extinction was limited to the deep-sea. At particular times in the geologic past, broad sectors of the Earth's environmental system, such as the global ocean, were susceptible to major reorganizations on short time scales. Brief (10² to 10³ yr), intense paleoenvironmental events can have large effects on the course of biotic evolution. At the end of the Paleocene, global climate change crossed a critical threshold, causing instability and mass extinction in the deep-sea.

The potential for the early Paleogene ocean to shift between high and middle-latitude dominated deep water production resulted from low vertical and meridional temperature gradients and multiple major sources of deep waters. Later in the Cenozoic, meridional and vertical temperature gradients strengthened, leading to decreased opportunities for such drastic switches in deep water sources. Thus, it is unlikely that the oceanographic changes and associated extinction event that occurred at the end of the Paleocene would have been repeated during the middle to late Cenozoic. The historic record reveals no such change.

Acknowledgments

This contribution was supported by the National Science Foundation (Division of Polar Programs) DPP-9218720 to J.P.K. and (Ocean Sciences) OCE 9101662 to L.D.S.

References

• Alvarez, L. W., W. Alvarez, F. Asaro, and H. V. Michel (1980). Extraterrestrial cause for the Cretaceous-Tertiary extinction, Science 208 , 1095-1108. [PubMed: 17783054]
• Barron, E. J. (1987). Eocene equator-to-pole surface ocean temperatures: A significant climate problem? Paleoceanography 2 , 729-739.
• Beckmann, J. P. (1960). Distribution of benthonic foraminifera at the Cretaceous-Tertiary boundary of Trinidad (West Indies), in Report of the 21 st Session, Norden, Part V, The Cretaceous-Tertiary Boundary , International Geological Congress, pp.57-69.
• Benson, R. H. (1979). In search of lost oceans: A paradox in discovery, in Historical Biogeography, Plate Tectonics, and the Changing Environment , J. A. Gray, editor; and A. J. Boucot, editor. , eds., Oregon State University Press, Eugene, pp.379-389.
• Berggren, W. A., and K. G. Miller (1989). Cenozoic bathyal and abyssal calcareous benthic foraminiferal zonation, Micropaleontology 35 , 308-320.
• Berggren, W. A., D. V. Kent, J. J. Flynn, and J. A. Van Couvering (1985). Cenozoic geochronology, Geological Society of America Bulletin 96 , 1407-1418.
• Bernard, J. M. (1986). Characteristic assemblages and morphologies of benthic foraminifera from anoxic, organic-rich deposits; Jurassic through Holocene, Journal of Foram. Research 16 , 207-215.
• Braga, G., R. De Biase, A. Grunig, and F. Proto-Decima (1975). Foraminiferi bentonici del Paleocene e dell'Eocene della Sezione Possagno, Schweizerische Palaeontologische Abhandlungen 97 , 85-111.
• Brass, G. W., J. R. Southam, and W. H. Peterson (1982). Warm saline bottom water in the ancient ocean, Nature 196 , 620-623.
• Broecker, W. S., and T.-H. Peng (1982). Tracers in the Sea , Eldigio Press, Palisades, N.Y., 690 pp.
• Cande, S. C., and D. V. Kent (1992). A new geomagnetic polarity time scale for the Late Cretaceous and Cenozoic, Journal of Geophysical Research 97 (B 10), 13,917-13,951.
• Case, J. A. (1988). Paleogene floras from Seymour Island, Antarctic Peninsula, in Geology and Paleontology of Seymour Island, Antarctic Peninsula , Geological Society of America Memoir 169, pp.523-530.
• Childress, J. J. (1983). Oceanic biology: Lost in space? in Oceanography: The Present and Future , P. Brewer, editor. , ed., Springer, New York, pp.127-135.
• Corfield, R. M. (1987). Patterns of evolution in Palaeocene and Eocene planktonic foraminifera, in Micropalaeontology of Carbonate Environments , M. M. Hart, editor. , ed., Eflis Horwood, Chichester, 93-110.
• Corliss, G. H., and C. Chen (1988). Morphotype patterns of Norwegian deep-sea benthic foraminifera and ecological implications. Geology 16 , 716-719.
• Covey, C., and E. J. Barron (1988). The role of oceanic heat transport in climatic change, Earth Science Reviews 24 , 429-455.
• Dercourt, J., et al. (1986). Geological evolution of the Tethys belt from the Atlantic to the Pamir since the Lias, Tectonophysics 123 , 241-315.
• Flessa, K. W. (1990). The "facts" of mass extinctions, Geological Society of America Special Paper 247, 1-7.
• Hallam, A. (1989). Catastrophism in geology, in Catastrophes and Evolution , S. V. M. Clube, editor. , ed., Cambridge University Press, pp.25-55.
• Haq, B. U. (1981). Paleogene paleoceanography—Early Cenozoic oceans revisited, Oceanologica Acta 4 (Suppl. to Vol. 4) , Proceedings of the 26^th International Geological Congress, pp.71-82.
• Hay, W. W. (1989). Paleoceanography: A review for GSA centennial, Geological Society of America Bulletin 100 , 1934-1956.
• Hovan, S. A., and D. K. Rea (1992). Paleocene/Eocene boundary changes in atmospheric and oceanic circulation: A Southern Hemisphere record, Geology 20 , 15-18.
• Janecek, T. R., and D. K. Rea (1983). Eolian deposition in the northeast Pacific: Cenozoic history of atmospheric circulation, Geological Society of America Bulletin 94 , 730-738.
• Kaiho, K. (1988). Uppermost Cretaceous to Paleogene bathyal benthic foraminiferal biostratigraphy of Japan and New Zealand: Latest Paleocene-Middle Eocene benthic foraminiferal species turnover, Reviews Paléobiol. , Vol. Spec. 2 , 553-559.
• Katz, M. E., and K. G. Miller (1993). Early Paleogene benthic foraminiferal assemblage and stable isotope composition in the Southern Ocean, ODP Leg 114, in Proceedings of the Ocean Drilling Program, Scientific Results 114 , P. F. Ciesielski, editor; and Y. Kristoffersen, editor. , eds., Ocean Drilling Program, College Station, Texas.
• Kennett, J. P. (1977). Cenozoic evolution of Antarctic glaciation, the circumAntarctic Ocean, and their impact on global paleoceanography, Journal of Geophysical Research 82, 3843-3860.
• Kennett, J. P., and P. F. Barker (1990). Latest Cretaceous to Cenozoic climate and oceanographic developments in the Weddell Sea, Antarctica: An ocean-drilling perspective, in Proceedings of the Ocean Drilling Program, Scientific Results 113 , P. F. Barker, editor; , J. P. Kennett, editor. , et al., eds., Ocean Drilling Program, College Station, Texas, pp.937-960.
• Kennett, J. P., and N. J. Shackleton (1976). Oxygen isotopic evidence for the development of the psychrosphere 38 Myr. ago, Nature 260 , 513-515.
• Kennett, J. P., and L. D. Stott (1990. a). Latest Cretaceous to Cenozoic climate and oceanographic developments in the Weddell Sea, Antarctica: An ocean-drilling perspective, in Proceedings of the Ocean Drilling Program, Scientific Results 113 , P. F. Barker, editor; , J. P. Kennett, editor. , et al., eds., Ocean Drilling Program, College Station, Texas, pp.865-880.
• Kennett, J. P., and L. D. Stott (1990. b). Proteus and Pro-oceanus: Ancestral Paleogene oceans as revealed from Antarctic stable isotopic results; ODP Leg 113, in Proceedings of the Ocean Drilling Program, Scientific Results 113 , P. F. Barker, editor; , J. P. Kennett, editor. , et al., eds., Ocean Drilling Program, College Station, Texas.
• Kennett, J. P., and L. D. Stott (1991). Abrupt deep-sea warming, palaeoceanographic changes and benthic extinctions at the end of the Palaeocene, Nature 353 , 225-229.
• Koch, P. L., J. C. Zachos, and P. D. Gingerich (1992). Correlation between isotope records in marine and continental carbon reservoirs near the Palaeocene/Eocene boundary, Nature 358 , 319-322.
• Lu, G., and G. Keller (1993) Climatic and oceanographic events across the Paleocene-Eocene transition in the Antarctic Indian Ocean: Inference from planktic foraminifera, Marine Micropaleontology .
• Mead, G. A., D. A. Hodell, and P. F. Ciesielski (1993). Late Eocene to Oligocene vertical oxygen isotopic gradients in the South Atlantic: Implications for warm saline deep water, in The Antarctic Paleoenvironment: A Perspective on Global Change 2 , J. P. Kennett, editor; and D. A. Warnke, editor. , eds., Antarctic Research Series, American Geophysical Union, Washington, D.C.
• Miller, K. G. (1991). The Paleocene/Eocene boundary in the context of Paleogene global climate change. Abstract, Geological Society of America Annual Meeting , A141.
• Miller, K. G., T. R Janecek, M. E. Katz, and D. J. Keil (1987). Abyssal circulation and benthic foraminiferal changes near the Paleocene/Eocene boundary, Paleoceanography 2(6), 741-761.
• Miller, K. G., M. E. Katz, and W. A. Berggren (1992). Cenozoic deep-sea benthic foraminifera: A tale of three turnovers, in Benthos '90, Studies in Benthic Foraminifera , Tokai University Press, pp.67-75.
• Nomura, R. (1991). Paleoceanography of upper Maastrichtian to Eocene benthic foraminiferal assemblages at ODP sites 752, 753 and 754, eastern Indian Ocean, in Proceedings of the Ocean Drilling Program, Scientific Results 121 , J. W. Pierce, editor. et al., eds., Ocean Drilling Program, College Station, Texas, pp.3-30.
• Oberhänsli, H., and K. J. Hsü (1986). Paleocene-Eocene paleoceanography, in Mesozoic and Cenozoic Oceans , K. J. Hsü, editor. , ed., American Geophysical Union Geodynamics Series 15, Washington, D.C., pp.85-200.
• Pak, D. K., and K. G. Miller (1992). Paleocene to Eocene benthic foraminiferal isotopes and assemblages: Implications for deepwater circulation, Paleoceanography 7(4), 405-422.
• Pospichal, J. M., and S. W. Wise, Jr. (1990). Paleocene to Middle Eocene calcareous nannofossils of ODP sites 689 and 690, Maud Rise, Weddell Sea, in Proceedings of the Ocean Drilling Program, Scientific Results 113 , P. F. Barker, editor; , J. P. Kennett, editor. , et al., eds., Ocean Drilling Program, College Station, Texas, pp.613-638.
• Premoli-Silva, I., and A. Boersma (1984). Atlantic Eocene planktonic foraminiferal historical biogeographic and paleohydrologic indices, Palaeogeography, Palaeoclimatology, Palaeoecology 67 , 315-356.
• Raup, D. M., and J. J. Sepkoski, Jr. (1984). Periodicity of extinctions in the geologic past, Proceedings of the National Academy of Sciences USA 81 , 801-805. [PMC free article: PMC344925] [PubMed: 6583680]
• Rea, D. K., J. C. Zachos, R. M. Owen, and P. D. Gingerich (1990). Global change at the Paleocene-Eocene boundary: Climatic and evolutionary consequences of tectonic events, Palaeogeography, Palaeoclimatology, Palaeoecology 79 , 117-128.
• Reid, J. L. (1979). On the contribution of Mediterranean Sea outflow to the Norwegian-Greenland Sea, Deep-Sea Research 26 , 199-223.
• Robert, C., and J. P. Kennett (1992). Paleocene and Eocene kaolinite distribution in the South Atlantic and Southern Ocean: Antarctic climatic and paleoceanographic implications, Marine Geology 103 , 99-110.
• Schlanger, S. O., and H. C. Jenkyns (1976). Cretaceous oceanic anoxic events: Causes and consequences, Geologie en Mijnbouw 55 , 179-84.
• Schnitker, D. (1979). Cenozoic deep-water benthic foraminifers, Bay of Biscay, in Initial Reports of the Deep-Sea Drilling Program 48 , L. Montadert, editor. et al., eds., U.S. Government Printing Office, Washington, D.C., pp.377-414.
• Shackleton, N. J. (1986). Paleogene stable isotope events, Palaeogeography, Palaeoclimatology, Palaeoecology 57 , 91-102.
• Shackleton, N. J., and M. A. Hall (1984). Carbon isotope data from Leg 74 sediments, in Initial Reports of the Deep-Sea Drilling Program 74 , U.S. Government Printing Office, Washington, D.C., pp.613-619.
• Shackleton, N. J., and J. P. Kennett (1975). Paleotemperature history of the Cenozoic and the initiation of Antarctic glaciation: Oxygen and carbon isotope analyses in DSDP sites 277, 279 and 281, in Initial Reports of the Deep-Sea Drilling Program 29 , J. P. Kennett, editor; , R. E. Houtz, editor. , et al., eds., U.S. Government Printing Office, Washington, D.C., pp.143-756.
• Shackleton, N. J., R. M. Corfield, and M. A. Hall (1985). Stable isotope data and the ontogeny of Paleocene planktic foraminfera, Journal of Foraminiferal Research 15 , 321-336.
• Sinha, A. and L. D. Stott (1994). The transfer of ¹³C change from the ocean to the atmosphere and terrestrial biosphere across the Paleocene/Eocene boundary: Criteria for terrestrial-marine correlations, in Early Paleogene Correlation in NW Europe , R. Knox, editor. , ed., Special Publication of the Geological Society, London: (in press).
• Stanley, S. M. (1984). Mass extinctions in the ocean, Scientific American 250 , 64-72.
• Stanley, S. M. (1987). Extinction , Scientific American Library, W. H. Freeman, New York.
• Stott, L. D. (1992). Higher temperatures and lower oceanic PCO₂: A climate enigma at the end of the Paleocene epoch, Paleoceanography 7(4), 395-404.
• Stott, L. D., and J. P. Kennett (1990). Antarctic Paleogene planktonic foraminifer biostratigraphy: ODP Leg 113, Sites 689 and 690, in Proceedings of the Ocean Drilling Program, Scientific Results 113 , P. F. Barker, editor; , J. P. Kennett, editor. , et al., eds., Ocean Drilling Program, College Station, Texas, pp.548-569.
• Stott, L. D., J. P. Kennett, N. J. Shackleton, and R. M. Corfield (1990). The evolution of Antarctic surface waters during the Paleogene: Inferences from the stable isotopic composition of planktonic foraminifera, ODP Leg 113, in Proceedings of the Ocean Drilling Program, Scientific Results 113 , P. F. Barker, editor; , J. P. Kennett, editor. , et al., eds., Ocean Drilling Program, College Station, Texas, pp.849-864.
• Thomas, E. (1989). Development of Cenozoic deep-sea benthic foraminiferal faunas in Antarctic waters, in Origins and Evolution of the Antarctic Biota , J. A. Crame, editor. , ed., Geological Society Special Publication, London: 47, pp.283-296.
• Thomas, E. (1990). Late Cretaceous-Early Eocene mass extinctions in the deep-sea, Geological Society of America Special Paper 247 , 481-495.
• Thomas, E. (1992). Cenozoic deep-sea circulation: Evidence from deep-sea benthic foraminifera, in The Antarctic Paleoenvironment: A Perspective on Global Change , Antarctic Research Series 56, American Geophysical Union, Washington, D.C., pp.141-165.
• Tjalsma, R. C., and G. P. Lohmann (1983). Paleocene-Eocene bathyal and abyssal benthic foraminifera from the Atlantic Ocean, Micropaleontology Special Publication 4 , Micropaleontology Press, New York, 94 pp.
• von Hildebrandt, A. (1962). Akad. Wiss. (Wien), Math.-Naturw. Klasse, Abh. n. ser. 108 , 1-182.
• Zachos, J., D. Rea, K. Seto, R., Nomura, and N. Niitsuma (1993. a). Paleogene and early Neogene deepwater paleoceanography of the Indian Ocean as determined from benthic foraminifera stable isotope records, in The Indian Ocean: A Synthesis of Results from the Ocean Drilling Program , R.A. Duncan, editor. et al., eds., Geophysical Monograph Series, American Geophysical Union, Washington, D.C.
• Zachos, J. C., K. C. Lohmann, J. C. G. Walker, and S. W. Wise (1993. b). Abrupt climate change and transient climates during the Paleogene: A marine perspective, Journal of Geology 101 , 193-215. [PubMed: 11537739]