1.2.1 The morphology, sculpture, and structure of the test of planktonic foraminifera

Foraminiferal tests rarely consist of only one chamber; usually, as the organism grows, it adds successively additional, progressively larger chambers to produce a test of varying complexity. The intrinsic buoyancy of the planktonic foraminifera is provided by the generally globular nature of their chambers. Some living planktonic foraminifera add a new chamber every day and grow at a rate that sees them increase their diameter by about 25% per day (Anderson and Faber, 1984; Bé et al., 1982; Caron et al., 1981; Erez, 1983; Hemleben et al., 1989).

Planktonic foraminifera have different patterns of chamber disposition (see Fig. 1.7):

• Trochospiral growth has the chambers coiling along the growth axis while also diverging away from the axis. The test has dissimilar evolute spiral and involute umbilical sides (Fig. 1.7A–C, H).

• Involute trochospiral growth has the chambers biserial or triserial in early stages, later becoming enrolled biserial, but with biseries coiled into a tight, involute trochospire (Fig. 1.7E).

• Planispiral growth has the chambers coiling along the growth axis but showing no divergence away from the axis. The test is biumbilicate, with both the spiral and umbilical sides of the test being identical and symmetrical relative to the plane of bilateral symmetry (Fig. 1.7G, I).

• Streptospiral growth has the chambers coiling in successively changing planes, or with the last globular chamber completely embracing the umbilical side (Fig. 1.7F).

• Uniserial, biserial, triserial multiserial, etc., patterns of growth have (after an initial planispiral or trochospiral stage) chambers arranged in one, two, three, or more rows in a regularly superposed sequence. The biserial form is planar (Fig. 1.7J, K), but multiserial forms can be three dimensions forming a conical test (Fig. 1.7L).

The planktonic foraminifera have a simple test with no internal structures and are, therefore, quite distinct from the larger benthic foraminifera. These latter can develop canal systems within the walls (Fig. 1.8A), plugs and pillars within the septa and umbilici, and internal toothplates that modify the routes of exit and ingress of the cytoplasm through the aperture (Fig. 1.8B, Ca; see BouDagher-Fadel, 2008). As can be seen in thin section, planktonic foraminifera do not develop plugs, pillars, or canal systems (see Fig. 1.8Cb–F).

The aperture of the planktonic foraminifera is the main opening of the last chamber cavity into the ambient environment. It can open completely in the umbilicus, umbilical/intraumbilical aperture (Fig. 1.9H), extend from the umbilicus toward the periphery of the test, intra–extraumbilical (Fig. 1.9K), or open completely over the periphery, unconnected to the umbilicus, extraumbilical (Fig. 1.9L). It may also be modified exteriorly by the development of an apertural tooth (inward projection(s) of the inner portion of the chamber wall into the aperture, see Fig. 1.9D), lips (Fig. 1.9F), a tegillum or tegilla (Fig. 1.9B, C), or a porticus or portici (Fig. 1.9A, E, I). However, the umbilical plates, such as the tegillum and portici, are similar to those of the benthic Rotaliina (e.g., Haynesina, Rosalina), which partially enclose the umbilical digestive cytoplasm with similar skeletal material (Alexander, 1985). The latter seems to be advantageous for extrathalamous digestion of disaggregated particles. As will be discussed in Chapter 4, in the Cretaceous, advanced forms of the Hedbergellidae evolved these analogous structures partially to enclose their umbilici (e.g., Ticinella, Rotalipora; see Fig. 4.9). By analogy with the benthic forms, therefore, this could have enabled a similar partial enclosure of an umbilical digestive ‘reservoir’ of cytoplasm to facilitate the effective ingestion of absorbed particles. Even predatory, carnivorous Cenozoic globigerinids partially digested disaggregated prey in extrathalamous digestion vacuoles (Spindler et al., 1984). In contrast to the Hedbergellidae, however, other Cretaceous forms, such as the Praehedbergellidae and the Schackoinidae, never developed extended portici.

In the Cenozoic, chamber growth developed so that some forms exhibited supplementary apertures (such as in Globigerinoides, Fig. 1.10B) or areal apertures (as in Orbulina, Fig. 1.10C). Occasionally, the aperture of a planktonic foraminifera is covered by a skeletal structure, a bulla (Plate 4.1, Fig. 18; Fig. 1.10D), or sometimes only a trace of it can be seen. Bullae extend over the umbilicus of the ultimate whorl and cover the primary, main, or supplementary apertures. They may also have marginal accessory apertures. These occurred in the Mesozoic and Cenozoic (see Chapters 4 and 5). However, the bullae of the Mesozoic differ from those of the Cenozoic species in often possessing short discontinuous ridges formed from pseudomuricae (see below), while those of the Cenozoic are perforate, not muricate (see below), and from one to four accessory, infralaminal apertures (e.g., Catapsydrax, Fig. 1.10D). In life, bullae would have allowed contact with the exterior through the accessory, infralaminal apertures at their margins. They too may have covered the umbilicus to conceal a mass of extrathalamous digestive cytoplasm by analogy with what is seen in some living benthic rotaliines (Alexander and Banner, 1984). It is possible that with the extraumbilical extension of the aperture and narrowing of the umbilicus, bullae became unnecessary.

The secreted calcitic or aragonitic tests of the planktonic foraminifera are always perforated by multitudinous small holes (Fig. 1.7). These perforations did not, however, form open connections between the cell within the test to the surrounding sea water, rather they enable the internal cytoplasm to make biochemical contact with a cytoplasmic sheet on the outer surface of the test, the extrathalamous cytoplasm (see Plate 1.1 below). They are closed off internally by the inner organic lining, allowing gas and solute exchange between intrathalamous and extrathalamous cytoplasm. Only small molecules and nutritional salts can penetrate via the perforation. They are also important for the nutrition of any algal endosymbionts hosted by the foram.

These perforations vary in size, with diameters between 0.28 and 2.5 μm in the microperforate forms (see Figs. 1.10E and 1.11A), and between 2.5 and 10 μm in the macroperforate forms (see Fig. 1.11B). They are abundant in the walls of each chamber of the test of, for example, all genera of the Globigerinida (e.g., Fig. 1.10A). However, microperforations are usually irregularly dispersed, while macroperforations are dense (see Fig. 1.11). Perforations must be distinguished from the much larger pores/areal apertures, which occur in the final chambers of some Cenozoic taxa, and which allow direct ingress and egress of cytoplasmic strands (pseudopodia) to and from the test limits. Pores occur commonly in the walls of the Miocene genera such as Praeorbulina, Orbulina (see Fig. 1.12D), and Globigerinatella, where they replace the primary aperture of ancestral forms (see Chapter 6; Plate 6.11, Fig. 1). Pore diameter appears to be directly related to environmental temperature (Bé, 1968), and test porosity is relatively uniform for species coexisting in the same latitudinal belts (Hemleben et al., 1989).

A variety of surface ornamentations also characterize planktonic foraminifera species, and these include perforation cones, pseudomuricae, muricae, short ridges or costellae, favose reticulations, pustules, muricocarinae or keels, spines, or smooth surfaces bearing a thin veneer of highly uniform calcite.

Perforation cones are conical-like structures (see Figs. 1.12A and 1.18A; and BouDagher-Fadel et al., 1997) with axial vents, which were built by extrathalamous cytoplasm over the perforations of many taxa of the Praehedbergellidae and the Heterohelicidae (Guembelitria, Plate 4.19, Figs. 1–3; Pseudotextularia, Plate 4.18, Figs. 22–28). Perforation cones are not “pore cones” (as they have been called by many authors, especially referring to their occurrence in the Heterohelicoidea), because they have nothing to do with pores (cf. perforations). The perforation cones on the surface of the test of Praehedbergellidae make their tests heavier and thicker toward the aperture; they become much lighter on the later chambers and may be absent on the last. This was achieved by depositing a new surface lamella of calcite over the earlier test as each new chamber was formed. The penultimate chamber had just one extra lamella of calcite, but earlier chambers have more and more lamellae, thickening the test and strengthening the cones. As a consequence, this gives rise to the strongest, most prominent cones nearest to the aperture.

The perforations have been shown in many benthic rotalines to act as links between the cytoplasm that is inside each chamber (intrathalamous) and that part which forms a layer outside the test (extrathalamous). It is from the extrathalamous cytoplasm that the pseudopodia arise. They are referred to as “granular”, as they carry granules that include excreted particles being carried away from the test and captured organic particles being carried into the test for ingestion. When the captured particles reach the extrathalamous cytoplasm covering the test, they are dragged over the test surface to the umbilicus and aperture for ingestion and digestion. The pseudopodia and the rest of the extrathalamous cytoplasm must also be the means by which the protozoan gathers oxygen and excretes waste gases. Ions pass through the perforations into and out of the intrathalamous cytoplasm (Angell, 1967; Banner and Williams, 1973; BouDagher-Fadel et al., 1997; Sheehan and Banner, 1972). Microperforate tests were adequate to allow gas and cation exchange when planktonic foraminifera were living in well oxygenated surface waters of shelf seas and open oceans, but the increased porosity that was created by the of macroperforate tests enabled efficient gas and cation exchange in less well oxygenated subsurface waters. The macroperforate Hedbergellidae and their descendants, therefore, are likely to have been more capable of populating deeper, cooler, and less oxygenated oceanic waters than were their microperforate ancestors.

In the Mesozoic, many Globigerinida have walls with surfaces that possess mound-like structures located between the surface perforations. In the Late Cretaceous, these structures were conical and pointed and were often as tall or taller than they were wide (Fig. 1.13E); these features were named muricae by Blow (1979). Muricae first appeared on species of the globigerinid genus Hedbergella in the Late Aptian. Large perforations were also developed by these forms, enabling them to occupy deeper water niches, prior to evolution of muricae. Phylogenetically, the muricae appear gradually first near the aperture and then evolve to be near the periphery and sutures. Muricae can also fuse meridionally across the test surface at right angles to the periphery, to form a costellae, as in Rugoglobigerina (Fig. 1.13D, G) and related genera. Muricae were present on the tests of the Globigerinoidea up to the close of the Eocene (e.g., Morozovella, Fig. 1.14A, D; see Chapter 5). The muricae developed in forms belonging to the superfamily Globigerinoidea independently and at different stages. The muricae of Cretaceous and Paleogene forms are morphologically indistinguishable, yet because of the major End Cretaceous extinction event they must have developed independently one from another (see Chapters 4 and 5). In both cases, however, they presumably were covered by extrathalamous extensions of the cytoplasm and were capable of carrying and transporting vacuoles, which would have given them a competitive advantage, and might explain why they were able to spread faster than the smooth praehedbergellids in the Cretaceous and globanomalids, which died out in the early part of the Middle Eocene. The smooth, nonmuricate planktonic foraminifera would have had no skeletal control on the distribution of sites for pseudopodial extensions into the surrounding seawater. The pseudopodia would either have extended irregularly or would have occupied a volume approximating to that of a sphere. If the same volume is compressed into a disc, the surface area for prey capture or collection must be very greatly increased. Therefore, species which could extrude the pseudopodia via muricae with greater structure would have had a considerable advantage in the collection of available suspended particulate nutrients.

In the Globotruncanidae of the Late Cretaceous, the muricae fuse together, along the periphery of the test, to form a muricocarina or keel (e.g., in Globotruncana, Fig. 1.13F). This also occurred in the planispiral Planomalina (Plate 4.15, Figs. 23 and 24). Some globotruncanids possessed two muricocarinae, which can be close together (appressed) or separated with imperforate peripheral band in between them (see Fig. 1.15). The keel can extend into the cameral sutures of the dorsal side and can extend into the ventral sutures of each chamber on the umbilical side. All the Globotruncanidae are extinct, so it is impossible to observe the function of these structures. However, it has been noted that the carinae of some predatory, biconvex benthic Rotaliina (e.g., Amphistegina), form the foundation of the “take-off” of food-gathering pseudopodia (Banner, 1978). We can surmise that the double muricocarinae of Globotruncana (Fig. 1.16A) were also associated with nutrition. If fans of pseudo-podia radiated from the extrathalamous cytoplasmic layer, which covered the test, and arose from each muricocarina, then they would drag organic particles to the muricacarinae where they would become disaggregated. These disaggregated particles would then be channelled along the imperforate band to the terminal face and then to the umbilicus. This would mean that no food particles would impede the function of the perforations on the dorsal and ventral faces of the chambers but would ensure that simple cytoplasmic flow would direct the nutrient particles straight to the umbilical aperture for their ingestion.

In the Paleogene, this evolutionary trend was repeated, with muricae again fusing at the periphery to form keeled Morozovella (see Figs. 1.14A, D and 1.16B), while in the Neogene, the compressed test of the globorotaliids is similarly reinforced by peripheral thickening of the chamber to give a keel (Fig. 1.16C). However, in the globorotaliids, muricae are absent and the test is smooth, so the keel is formed by the collapse of the chamber wall along the test periphery, where it serves an architectural function of strengthening the test (Hemleben et al., 1989). In the early stages of the globorotaliids, the keel area is still covered by perforations (see Fig. 1.16C), while in older chambers, the keel is strengthened by additional calcite layers (Hemleben et al., 1989).

The surface of the test in most nonspinose Neogene species bears small rounded calcitic knobs which have been called pustules (emended by Blow, 1979). They look similar to the perforation cones of the Praehedbergellidae or the muricae of the Hedbergellidae and their related forms and are also concentrated on the test surface in the vicinity of the aperture (s) and the umbilicus as exemplified in Turborotalia cerroazulensis (Fig. 1.17A) and Globorotalia margaritae (Fig. 1.17B). Here, their function seems to be the disaggregation, by rasping, of food particles prior to their ingestion through the apertures (Alexander and Banner, 1984) or digestion in the umbilical cytoplasmic “reservoir” (Alexander, 1985). They also may have served as anchor points for masses of rhizopodia (Hemleben, 1975; Hemleben et al., 1989). In the Paleogene, microperforate taxa can also possess pustules (e.g., Globastica/Globoconusa of the Danian, and the Globigerinitidae of the Late Palaeogene to Holocene; Plate 5.3, Figs. 1–5; Plate 6.1, Figs. 8–13; Chapters 5 and 6).

The Favuselloidea of the Jurassic and Early Cretaceous possessed neither pointed muricae nor blunt, scattered pustules. Instead, many of their tests possessed narrow, blunt or pointed, scattered projections which were more elevated and regularly spaced than pustules. BouDagher-Fadel et al. (1997) called these structures pseudomuricae. Pseudomuricae are often hollow (Samson et al., 1992; see Chapters 3 and 4; Figs. 1.13A and 1.18A) as muricae can be (Blow, 1979, plate 208), but never fuse peripherally into muricocarinae. However, in contrast, they can fuse into discontinuous ridges (e.g., as on Globuligerina, see Fig. 1.13 A), which are at random angles to each other, and which eventually themselves fuse together into favose reticulations (Fig. 1.13B; see Chapter 4), which appear to be solid (Fig. 1.18E). Muricae never fuse to form reticulations but occur between the macroperforations of the taxa which have them. The ridges of Favusella may still retain pseudomuricae at the junctions between the ridges of the reticulum (Fig. 1.13B). Pseudomuricae can enclose patches of microperforations, unlike anything seen in macroperforate forms (Fig. 1.18A).

True spines really only came to full development in the Eocene, yet by the end of the Paleogene spinose taxa had replaced muricate forms entirely (Fig. 1.19). Spines are very long, needle-shaped (aciculate) structures, circular or triangular in cross section (Hemleben et al., 1989, pp. 209–210), which have sub-cylindrical bases embedded in the chamber wall. Their morphology includes five general types (Hemleben et al., 1989; Saito et al., 1976):

• the Globigerina-type, round in cross section throughout (Fig. 1.19D);

• the Globigerinoides- and Orbulina-type, with round spines, but which may become triradiate distally (Fig. 1.19B);

• the Orcadia-type triangular spines throughout (Fig. 1.19A);

• the Globigerinella-type with triangular thin section at the base becoming triradiate (Plate 6.4, Fig. 16);

• the Hastigerina-type which are clearly triradiate from their base (Fig. 1.19H).

Clearly, spines could have a greater variety of function than the muricae of their ancestral taxa. Spines might have a dense or sparse distribution. In Fig. 1.19, one can note the different distribution of the spine bases found on extinct Globigerina (Fig. 1.19D) and Orcadia (Fig. 1.19A) from the recent seas. The spine bases, which become embedded in the ridges (Fig. 1.12C) between the macroperforations of post-Paleogene taxa, have been well studied in SEM by Bé (1969), Bé et al. (1973), Bé and Hemleben (1970), and Bé (1980). The spine bases are often prominent between perforation pits as in Clavatorella, or rarely prominent, and often obscure, as in Globorotalia (see Fig. 1.20).

It is known that the spines of living Globigerinidae are covered by extrathalamous extensions of the cytoplasm, capable not only of carrying vacuoles and transporting symbionts but also of performing the feeding and ion exchange functions (Adshead, 1980). They can ingest food particles from as far as possible from the host test and reject waste matter from similar distances (Hemleben et al., 1989). This is important particularly in turbulent shallow water where they are found. Spinose globigerinid species occupy shallow (mixed layer and intermediate depths, 0–100 m) planktonic habitats (Fairbanks et al., 1982), and harbour algal symbionts within their cytoplasm (Bé, 1982; Boltovskoy and Wright, 1976; Hemleben et al., 1989; Lipps, 1979; Murray, 1991). The presence of spines has been suggested to being linked to maintaining control over the depth of the habitat and to help the organism resist sinking through the water column (Bé, 1982; Hutchinson 1967; Lipps 1979). However, the presence or absence of spines cannot be the sole determining factor here, as there are a number of nonspinose planktonic foraminiferal species that occupy shallow-intermediate planktonic habitats too (e.g., see Figs. 5.22 and 6.21). An alternative hypothesis for the function of spines is that they may provide the organism with a metabolically inexpensive means of greatly increasing the area of its pseudopodial network, and in so doing, increase the opportunity for that network to capture food particles (Murray, 1991). Neither the acquisition of spines nor the ability to harbour algal symbionts can be regarded as an adaption to a specific depth habitat. Spines were developed by the globigerinid lineage well after the transition to a shallow-intermediate depth habitat had been made by their ancestral species (Macleod, 2001). Spines, however, also may have been an adaptation to assist in dispersing the symbiotic algae in such a way as to optimize diffusion of CO₂ and oxygen.