Acarinina multicamerata n. sp. (Foraminifera): a new marker for the Paleocene–Eocene thermal maximum

During the Paleocene–Eocene thermal maximum (PETM), low to mid-latitude planktic foraminiferal assemblages were characterized by blooms of the surface-dwelling Acarinina. Among this group a new ‘excursion taxon’ is identified, Acarinina multicamerata n. sp. Previously, this taxon was lumped together with Acarinina sibaiyaensis El-Naggar. Considering that A. sibaiyaensis already occurred prior to the hyperthermal event, both in open ocean and ocean margin deposits, it is proposed that these taxa are differentiated in order to avoid taxonomic and biostratigraphic ambiguities. Acarinina multicamerata n. sp. occurred exclusively during the PETM, hence this taxon represents an excellent biostrati-graphic marker of the PETM, while its common occurrence in various marine settings makes it an excellent marker of Subzone P5b or its new equivalent zone E1.


INTRODUCTION
The Paleocene-Eocene thermal maximum (PETM) represents a period of extreme global warmth (Zachos et al., 1993), associated with various biotic turnovers on land and in the sea. Among the latter is an evolutionary rejuvenation in planktic foraminifera, leading to short-lived species of Acarinina and Morozovella (Kelly et al., 1996). During the PETM, planktic foraminiferal assemblages in open-marine environments worldwide are dominated strongly by Acarinina. Both in open ocean ODP sites and in the marginal Tethys, a number of authors have described the occurrence of new planktic foraminifera taxa, called 'excursion' taxa, at the PETM; these include Acarinina sibaiyaensis, A. africana and Morozovella allisonensis. These are generally thought to have evolved during the early part of the PETM (Kelly et al., 1996(Kelly et al., , 1998Bolle et al., 2000;Ouda et al., 2003). Observations on Egyptian successions, however, show that A. sibaiyaensis and A. africana (=M. africana) indeed bloom during the PETM, but had already evolved prior to this event (Speijer et al., 2000;Guasti & Speijer, 2007), confirming the early observations of El-Naggar (1966), the author who erected these taxa.
This study examined material from upper Paleocene-lower Eocene successions in Egypt (Dababiya, Gebel Duwi, Gebel Aweina, Gebel Qreiya and Wadi Nukhl) and Jordan (Gebel Qurtassyat) (Figs 1, 2) in order to investigate the occurrence and taxonomy of excursion taxa. It was found that there are some ambiguities in the taxonomy of the PETM excursion taxa. In this paper, the longer-ranging species A. sibaiyaensis is differentiated from a multi-chambered Acarinina species, which only occurred during the PETM. The latter is defined as A. multicamerata n. sp. In addition. environmental scanning electron microscope (ESEM) images of the holotypes of A. sibaiyaensis and A. africana are provided. Images of these holotypes, as well as the holotype and paratype of A. multicamerata n. sp., were taken at the Natural History Museum of London (UK), using an ESEM with variable pressure (LEO 1455 VP), which allowed photography of specimens in a mounting slide without coating.

ACARININA SIBAIYAENSIS
The holotypes of Acarinina sibaiyaensis and A. africana were described originally by El-Naggar (1966) as Globorotalia sibaiyaensis and Globorotalia africana, respectively (Pl. 1). He observed these species in the upper Paleocene, at Gebel Aweina in the Nile Valley, Egypt. In the original description, Globorotalia sibaiyaensis was described as a compressed small, globular species, in which the last whorl is composed of 5.5 chambers, having a long narrow aperture. El-Naggar (1966) suggested that this taxon may have evolved from G. perclara Loeblich & Tappan (1957).
Our observations uncovered a number of problems on the evolution, morphology and range of A. sibaiyaensis in the literature. For instance, Kelly et al. (1998)  different description of A. sibaiyaensis compared with the holotype description. These authors considered A. sibaiyaensis as a globular flat species with 6-9 chambers, which in their view seemed to evolve from Acarinina soldadoensis Brönniman (1952). To support this evolutionary linkage, Kelly et al. (1998) showed a morphological sequence from A. soldadoensis to A. sibaiyaensis ( fig. 9, p. 150). In the current authors' view, the specimen presented as A. soldadoensis (specimen 'A') in this cline seems already very close to A. sibaiyaensis sensu El-Naggar (1966) and quite different from A. soldadoensis. Acarinina soldadoensis is composed of 4-5 chambers gradually increasing in size, distinctively elongated in the direction of the coiling axis of the test; at the umbilical side the chambers tend to become pointed. The sutures of the spiral side are oblique, giving the impression of overlapping chambers (Olsson et al., 1999). These diagnostic features are missing in the specimens considered by Kelly et al. (1998) as A. soldadoensis. At the other end of the cline, the multi-chambered (6-9 chambers) forms of A. sibaiyaensis sensu Kelly et al. (1998) are readily distinguished from A. sibaiyaensis El-Naggar (1966), although the taxa may be linked phylogenetically. Moreover, A. sibaiyaensis should not be confused with Acarinina esnaensis LeRoy (1953), which also occurs in the Esna Formation during the Paleocene. Acarinina esnaensis differs from A. sibaiyaensis most notably in having a quadrate outline and only 4-4.5 subovate to subspherical chambers in the last whorl. Arenillas et al. (2004) also found specimens of the multichambered form of A. sibaiyaensis within the PETM horizons. These authors suggested that Acarinina strabocella Loeblich & Tappan (1957) could have been the ancestor of A. sibaiyaensis, although this would be in conflict with observations by various other authors, who recorded the highest occurrence of A. strabocella in lower Zone P4 (Lu et al., 1998;Olsson et al., 1999). . Acarinina multicamerata n. sp., holotype: 7, umbilical view; 8, side view; 9, spiral view; 10, 11, details of the wall texture (scale bar 10 µm). From sample BI3+8-12, 15 m above the base of the Esna Formation at Gebel Duwi, Red Sea Coast, Egypt. Collection number PF 67590, the NHM. Scale bar represents 100 µm, unless otherwise specified. Berggren et al. (2006) re-described A. sibaiyaensis based on material from a couple of ODP cores, most notably from Bass River, New Jersey. In their description these authors considerably widened the concept of A. sibaiyaensis by including specimens with much more numerous chambers to a whorl. In contrast, El-Naggar's holotype is comprised of 5.5 chambers in the last whorl, whereas one of his paratypes (depicted here in Pl. 2) has only 4 chambers in the last whorl. Apparently the specimens of El-Naggar came from an upper Paleocene population with relatively few chambers to a whorl. This is the kind of variation that was also observed for A. sibaiyaensis in upper Paleocene deposits in Egypt. Berggren et al. (2006) studied lowermost Eocene populations of flat-spired Acarinina and proposed intergradations between typical specimens of A. sibaiyaensis conformable with El-Naggar's concept and a variety with more numerous chambers in the last whorl. The latter variety has been observed associated with the PETM also by various other authors (i.e. Kelly et al., 1996Kelly et al., , 1998Pardo et al., 1999;Arenillas et al., 2004) and was also identified as A. sibaiyaensis.

provided a very
In the current authors' view the variety with 6-9 chambers in the last whorl qualifies as a new species, Acarinina multicamerata n. sp., described below. The concept of this taxon includes specimens 7, and 9-14 of plate 9.21 of Berggren et al. (2006). Besides the larger number of chambers in the last whorl, these specimens have a more rounded outline and a wider umbilicus compared with A. sibaiyaensis. All other specimens depicted by Berggren et al. (2006) seem to represent transitional forms between A. sibaiyaensis and A. multicamerata in having features typical of both these specimens. This reinforces the idea that during the PETM diversification of muricate taxa (Acarinina, Morozovella) occurred (Kelly et al., 1998) and that a variety of flat-spired Acarinina bloomed. Before the PETM, morphological variability was subdued and only the typical A. sibaiyaensis occurred sporadically. During the PETM A. multicamerata diverged from A. sibaiyaensis.
Lumping these two morphologically distinct taxa into one species would also lead to a loss of stratigraphic resolution. Acarinina multicamerata has been observed exclusively within PETM beds. A. sibaiyaensis and A. africana (by some considered as Morozovella because of the slightly keeled last chamber(s)) already appeared during the late Paleocene as indicated by the data of El-Naggar (1966). He recorded these taxa from the topmost metres of the upper Paleocene 'Middle Owaina Chalk' (=Tarawan Formation) and the lowermost 14 m of the overlying 'Upper Owaina Shale' (=Esna Formation) in the Aweina section (El-Naggar, 1966, fig. 18). This corresponds to a range from Zone P4 to the lower part of Zone P5 (Subzone P5a and possibly slightly higher) of Berggren et al. (1995), when compared to more recent studies on this section (Speijer et al., 1995(Speijer et al., , 2000Ouda et al., 2003). Observations have confirmed the occurrence of A. africana and A. sibaiyaensis in Aweina at least 40 cm below the P/E boundary (Pl. 2, Table 1). El-Naggar's observations lower in the Aweina section cannot be confirmed, because planktic taxa in samples from below the PETM (upper Tarawan Fm. and lower Esna Fm) are poor in number and preservation (Speijer & Schmitz, 1998). Ouda et al. (2003) also observed these taxa below the unconformity at the basis of the 'calcarenitic bed', which, according to earlier studies (Speijer et al., 1995(Speijer et al., , 2000Schmitz et al., 1997), marks the P/E boundary. In contrast to the opinion of Ouda et al. (2003), who considered this interval an early part of the PETM, the beds below the 'calcarenitic bed' belong to the uppermost Paleocene as indicated by the position of the carbon isotopic excursion coinciding with the unconformity (Speijer et al., 1995;Guasti & Speijer, 2007).
Derivation of name. The species name multicamerata derives from the numerous chambers in the final whorl, which characterize this taxon.

Diagnosis.
Test coiled in a low trochospire, containing 6-9 chambers in the final whorl. The chambers are globular and gradually increase in size. Dorsally flattened tests with rounded and lobate peripheral margin. The umbilicus is generally deep. The sutures are radial on both sides and strongly depressed, particularly on the ventral side. The aperture is an interiomarginal, extraumbilical-umbilical low elongate arch extending from the umbilicus to the periphery. Wall texture: non spinose, muricate.
Locality and horizon. Gebel Duwi, 26(05# N and 34(07# E, 12 km east of the town of Quseir (Red Sea Coast, Egypt). Sample BI3+8-12, collected from the 30 cm thick pink coprolite-rich marl bed, 15 m above the base of the Esna Formation. This bed appears to be an equivalent to Dababiya Quarry bed 3 in the GSSP section of Dababiya (Dupuis et al., 2003).

Remarks.
Acarinina multicamerata differs from A. sibaiyaensis in several features. A. multicamerata has a larger number of chambers per whorl and the outline is overall more rounded.
The chambers increase more gradually in size compared to A. sibaiyaensis. In some specimens, the wall texture of A. multicamerata is dominated by the large pores, instead of being covered with spiky pustules. The umbilicus is generally narrower in A. sibaiyaensis, resulting in a more tightly coiled test compared to A. multicamerata. The number of chambers (6), a rounded outline and a wider umbilicus are also evident in the specimens of Berggren et al. (2006, pl. 9.21, figs 9-14), suggesting the attribution of these specimens to A. multicamerata. All the other specimens seem to represent a transitional form between A. sibaiyaensis and A. multicamerata.
Origin of the species. El-Naggar suggested that A. sibaiyaensis evolved from G. perclara Loeblich & Tappan (1957), which already occurred in the Danian, but it has been reinterpreted as benthic foraminifera by Liu et al. (1998). Instead, Kelly et al. (1998) suggested that A. sibaiyaensis (=A. multicamerata) evolved from A. soldadoensis. Berggren et al. (2006) proposed Acarinina esnehensis as the ancestor of A. sibaiyaensis. The possibility that A. sibaiyaensis could derive from A. esnehensis is not excluded, although the morphology is distinctively different. It is suggested that A. multicamerata is derived from A. sibaiyaensis, as evidenced in the similar wall-textures and in the occurrence of transitional specimens. A. multicamerata diverged from the parent species to adapt to different environmental conditions.
Repository. The holotype (PF 67590) and three paratypes (PF 67591, PF 67592, PF 67593) are deposited at the Natural History Museum of London (UK).

CONCLUSION
The studied material indicates that Acarinina multicamerata is the only known Acarinina species occuring exclusively within the PETM beds. It can be distinguished easily from its precursor species A. sibaiyaensis, with which it has been confused previously. Hence, it is proposed that A. multicamerata n. sp. and M. allisonensis are presently the only true PETM excursion taxa among planktic foraminifera. Because of the exclusive and common occurrence within the PETM, A. multicamerata acts as an excellent marker of Subzone P5b or E1, replacing the less common M. allisonensis or the longer-ranging A. sibaiyaensis.