CENOZOIC CALCAREOUS NANNOPLANKTON CLASSIFICATION

Part 1, Heterococcoliths

Jeremy R. Young, Palaeontology Dept., NHM Cromwell Road, London, SW7 5BD, UK &

Paul R. Bown, Dept. of Geological Sciences, UCL, Gower Street, London, WC1E 6BT, UK

The original version of this ms was printed in the Journal of Nannoplankton Research, issue 19/1

Introduction

The purpose and philosophy of this contribution are explained in the introductory section (Young & Bown, above). As in the Mesozoic section (Bown & Young, 1997), a three-level order-family-genus classification is used, as far as seems reasonable, based on current knowledge. In addition, a set of informal numbered groupings (1. Murolith coccoliths to 5c. Nannoliths consisting of a single crystal-unit, and lacking radial symmetry) are used to provide a logical, but very possibly artificial, organisation, particularly of families and genera incertae sedis. For completeness, living coccolithophorids are included, even when they have no known fossil record, this part of the classification being largely derived from Jordan & Green (1994) and Jordan et al. (1995). Genera with extant species are indicated by an asterisk *. If no fossil representatives are known, a second asterisk is added **.

CONTENTS

CENOZOIC (Young & Bown 1997)

Order EIFFELLITHALES
Family CHIASTOZYGACEAE
Order ZYGODISCALES
Family HELICOSPHAERACEAE
Family PONTOSPHAERACEA
Family ZYGODISCACEAE

Order STEPHANOLITHIALES
Family CALCIOSOLENIACEAE
Order SYRACOSPHAERALES
Family SYRACOSPHAERACEAE
Order RHABDOSPHAERALES
Family RHABDOSPHAERACEAE

Order PRINSIALES
Family PRINSIACEAE
Family NOELAERHABDACEAE
Order COCCOSPHAERALES
Family COCCOLITHACEAE
Family CALCIDISCACEAE
Family PLEUROCHRYSIDACEAE
Order WATZNAUERIALES
Family WATZNAUERIACEAE

Family HYMENOMONADACEAE
Family PAPPOSPHAERACEAE
Genera incertae sedis

1. HETEROCOCCOLITHS

1.1. Murolith heterococcoliths

1.1a. Imbricating muroliths (loxoliths)

Order EIFFELLITHALES Rood, Hay & Barnard, 1971

This order is predominantly Mesozoic, see Mesozoic section for discussion.

Family CHIASTOZYGACEAE Rood, Hay & Barnard, 1973 emend. Varol & Girgis, 1994

The following two genera are known from both the Paleocene and Maastrichtian and display the typical Zeugrhabdotus-type rim-structure.

• Zeugrhabdotus Reinhardt, 1965 {The only Tertiary species, Z. sigmoides, has previously been included in Placozygus but shows typical Zeugrhabdotus rim and central-area structure}
• Neocrepidolithus Romein, 1979 {broad, high rim with narrow or closed central-area which may be spanned by bars}

Order ZYGODISCALES Young & Bown 1997

Description: Muroliths, and modified descendants, with an outer rim-cycle of V-units showing anticlockwise imbrication and an inner rim-cycle showing clockwise imbrication - the opposite imbrication sense to Eiffellithales. This is a diverse group but with clear evolutionary relationships (e.g. Romein, 1979; Aubry, 1989). Central-area structures include disjunct transverse bars, diagonal crosses and perforate plates but no spines.

Comments: Often assumed to have evolved from the Eiffellithales, via Chiastozygus, Zeugrhabdotus or Placozygus. However, this is not based upon any directly observable transitions and the opposite imbrication directions of the rim-cycles in the two orders makes this questionable. It is equally likely that the first member of the family, Neochiastozygus, evolved from a quite different Cretaceous ancestor.

N.B. Regroupings - a conventional subdivision into three families is followed here, however:

1. The Pontosphaeraceae vs. Zygodiscaceae subdivision is not obviously logical.

2. The Neococcolithes group arguably should be a separate group from the rest.

3. Grouping could be done via a new suborder or by making them all subfamilies of Pontosphaeraceae (this would have to be used since it has priority).

Family HELICOSPHAERACEAE Black, 1971

Description: Extant species are motile, forming ellipsoidal coccospheres with a prominent flagellar opening. Coccoliths are arranged spirally round the coccosphere and may vary slightly in size and shape from the antapex to the flagellar pole. Outer rim (V-units) of the coccolith is modified into a helical flange, ending in a wing or spike. R-units form the baseplate and ?extend to form a blanket of small elements. Central-area bars are conjunct, disjunct or absent.

Helicosphaera* Kamptner, 1954 (= Helicopontosphaera Hay & Mohler, 1967) {coccoliths with helical flange, subgroups can be recognised based on presence/absence of a disjunct bar, bar orientation, flange shape, etc.}

Family PONTOSPHAERACEAE Lemmermann, 1908

Description: Extant species apparently non-motile, coccospheres subspherical and may possess strongly-modified equatorial coccoliths (Scyphosphaera). V-units form narrow outer rim-cycle. R-units form inner rim, baseplate and ?blanket. No disjunct structures.

• Pontosphaera* Lohmann, 1902 (= Crassapontosphaera Boudreaux & Hay, 1969; Discolithina Loeblich & Tappan, 1963; Discolithus Huxley, 1868; KoczyiaBoudreaux & Hay, 1969) {central-area solid or with a variable number of pores}
• Scyphosphaera* Lohmann, 1902 {like Pontosphaera but with elevated equatorial coccoliths - lopadoliths [N.B.Calciopilleus and Tintinnabuliformis are lopadolith-like coccoliths with apparently different ultrastructures, see incertae sedis heterococcoliths]}
• Transversopontis Hay, Mohler & Wade, 1966 {central-area spanned by a conjunct bar, usually oblique}

Family ZYGODISCACEAE Hay & Mohler, 1967

A. Rim formed from well-developed V- and R-units, with opposite imbrication directions

• Jakubowskiella Varol, 1989 {open central-area}
• Lophodolithus Deflandre in Deflandre & Fert, 1954 {asymmetrical, often with disjunct bar}
• Neochiastozygus Perch-Nielsen, 1971 {with diagonal or asymmetric cross in central-area}
• Zygodiscus Bramlette & Sullivan, 1961 {symmetrical, with disjunct bar}

B. Rim formed from V-units; R-units vestigial or absent

Assignment of these genera to the Zygodiscaeae is based on imbrication direction of V-units, and putative evolutionary link from Neochiastozygus to Neococcolithes.

• Neococcolithes Sujkowski, 1931 (= HeliorthusBronnimann & Stradner, 1960; Indumentalithus Vekshina, 1959; Zygolithus Kamptner ex Matthes, 1956) {H-shaped cross in central-area}
• Chiphragmalithus Bramlette & Sullivan, 1961 {High wall and well-developed central-area cross}
• ? Genus Isthmolithus Deflandre, 1954 {parallelogram-shape, affinities to Zygodiscaceae uncertain, see Aubry (1988)}
• Nannotetrina Achuthan & Stradner, 1969 (= Nannotetraster Martini & Stradner 1960) {X-shaped cross with no rim, probably derived from Chiphragmalithus or Neococcolithes by loss of rim (Perch-Nielsen, 1985). N.B. Species were assigned to Nannotetraster until Acuthan & Stradner (1969) showed that it is a junior synonym of Micula}

1.1b. Other muroliths and planoliths

Comments: The structure of the three groups included here has not been fully worked out, and there is no direct fossil evidence as to their phylogenetic relationships. Affinities between them have been inferred on the basis of central-area structures which are characteristically composed of numerous concentric cycles of apparently disjunct elements, with tangential c-axis orientations (T-units). The outermost of these central-area cycles usually consists of radial lath-shaped elements which alternate around the rim with rim elements. This type of structure is shown by the three families included here and so it has been inferred that they have a common ancestry. However the rim structures are markedly different in the three so the apparent central-area similarities maybe misleading. All these groups are well known from the modern plankton where they typically have medium-sized coccospheres covered with large numbers of small coccoliths (often <3 µm). Many species show polymorphism. In the fossil record the small size of the coccoliths makes identification problematic.

Order STEPHANOLITHIALES Bown & Young 1997

Family CALCIOSOLENIACEAE Kamptner, 1937

Description: Extant species are motile with elongate fusiform coccospheres and spine-bearing polar coccoliths. Coccoliths are rhombic muroliths without flanges, usually termed scapholiths. The rim is predominantly formed of V-units, with small R-units at the base. The central-area has a single lath-cycle, i.e. bars are formed of two laths, one from each side of the central-area.

Comments: This family is not recognised in many classifications of the extant coccolithophores, with the genera instead being included in the Syracosphaeraceae, mainly due to similarities between central-area structures. We prefer to maintain it as a separate family since the rim structure is not like that of typical Syracosphaeraceae. The group certainly extends into the Mesozoic and may have evolved from the Stephanolithiaceae. In the LM, the rhombic shape makes it easy to recognise these coccoliths but they are too small to be identified to a lower level and many palaeontologists assign all scapholiths to the somewhat artificial species Scapholithus fossilis.

• Anoplosolenia* Deflandre, 1952 {no polar spines, coccoliths medium-sized - 4-7 µm}
• Calciosolenia* Gran, 1912 (= AcanthosoleniaBernard, 1939; ?= Scapholithus Deflandre, 1954) {with polar spines, coccoliths small - 2.5-3.5 µm}

Order SYRACOSPHAERALES Ostenfeld, 1899

Family SYRACOSPHAERACEAE Hay, 1977

Description: Extant species are motile, typically with elaborate coccospheres, often showing dithecatism (development of distinct inner and outer layers of coccoliths) and/or modified polar coccoliths. The endothecal (inner layer) coccoliths are normally relatively conservative in form, typically muroliths with the rim-structure described above, a well-developed central-area lath-cycle and variable inner central-area; they are often termed caneoliths. Exothecal coccoliths are much more variable, including discoidal and dome-shaped forms (cyrtoliths).

Comments: These coccoliths are typically delicate and only rarely preserved. The recent tendency (e.g. Jordan et al., 1995), pending a detailed revision, has been to combine the whole range of forms into the single genus Syracosphaera . Polar coccoliths are often only mildly dimorphic, slightly smaller, more angular in shape and bear larger spines. In other cases they are highly-modified, forming elaborate whorl structures. These forms are recognised as separate genera, as are a few other distinctive forms. The fossil record of the family is poor but extends back into the Paleogene; fossil specimens are normally assigned to Syracosphaera.

A. Genera with appendages

These genera have a whorl of appendages formed from highly modified coccoliths around either the apical (flagellar) or antapical pole. They are all monothecate and the body coccoliths are muroliths with a single, weak, proximal flange. This grouping is convenient but probably artificial and so we do not recommend describing a taxon based on it.

• Calciopappus** Gaarder & Ramsfjell, 1954 emend. Manton & Oates, 1983 {monothecate, with flangeless muroliths, apical coccoliths modified into elongate spines}
• Michaelsarsia** Gran, 1912 emend. Manton et al. , 1984 (= Halopappus Lohmann, 1912) {monothecate, with flangeless muroliths, apical appendages formed from a string of three highly-modified coccoliths - osteoliths}
• Ophiaster** Gran, 1912 emend. Manton et al.,1984 {monothecate, with flangeless muroliths, antapical appendages formed from a string of several highly modified coccoliths - osteoliths}

B. Genera without appendages

• Alisphaera** Heimdal, 1973 {monothecate, coccoliths are placolith-like, with asymmetrical flange bearing a spike or protrusion}
• Alveosphaera** Jordan & Young, 1990 {monothecate, coccoliths are elongate oblong muroliths, scapholith-like}
• Canistrolithus** Jordan & Chamberlain, 1993 {monothecate, coccoliths are elongate oblong muroliths, wall/rim weakly imbricate (anticlockwise) with distal flange}
• Coronosphaera* Gaarder in Gaarder & Heimdal, 1977 {monothecate, coccoliths are flangeless muroliths, with strongly imbricate (anticlockwise) rims; placement within this family is conventional but rim structure is anomalous}
• Syracosphaera* Lohmann, 1902 {usually dithecate, exothecal coccoliths highly variable, endothecal coccoliths include muroliths with 1, 2 or 3 flanges and placolith-like form, apical coccoliths often have spines}
• [Caneosphaera** Gaarder in Gaarder & Heimdal, 1977] {monothecate, coccoliths placolith-like, with asymmetrical flange; the included species are now usually recombined in Syracosphaera }
• [Deutschlandia* Lohmann, 1912] {exothecal coccoliths disc-shaped, the included species are now usually recombined in Syracosphaera}

Order RHABDOSPHAERALES Ostenfeld, 1899

Family RHABDOSPHAERACEAE Lemmermann, 1908

Description: Coccospheres may be motile or non-motile, typically they have spine-bearing and non-spine-bearing coccoliths with similar shields. The spine-bearing coccoliths may be confined to the poles or distributed around the coccosphere, greatly increasing its outer diameter. The coccoliths are disc-shaped (planoliths, see Young et al., in press) with a distinct, slightly elevated rim. In modern species, this is formed of two cycles of elements: a lower/inner cycle showing strong obliquity and an upper/outer cycle of simple non-imbricate elements (Kleijne, 1992). The upper/outer cycle is formed of V-units, the orientation of the inner/lower cycle is unclear. Central-area T-unit cycles are well developed, including both radial laths and usually a central spine or protrusion formed of numerous small elements with a spiral arrangement.

Comments: The spinose coccoliths are easy to spot in the LM and can usually be identified. In the Eocene, the Rhabdosphaeraceae form a diverse and abundant group (Perch-Nielsen, 1985; Varol, 1989; Shafik, 1989).

Eocene genera

Rhabdoliths are common in the Eocene and show broadly similar structures and morphologies to modern forms but detailed homologies with the extant genera are not clear. They are characterised by complex multicyclic shields and often multitiered central-structures. In addition, the outermost shield-cycle has far fewer elements than the inner shield-cycle. Forms with cap-shaped protrusions appear particularly complex and Shafik (1989) differentiated numerous genera on structural details, although these may prove to be oversplit (N.B. Shafik (1989) has priority over Varol (1989); official publication dates are March 1989 vs May 1989, NHM library accession dates are August 1989 vs June 1990).

• Blackites Hay & Towe, 1962 emend Stradner & Edwards, 1968; Varol, 1989 {multicyclic rhabdolith with hollow spine and flaring collar}
• Genera Cepekiella Roth, 1970; DiscoturbellaRoth, 1970; Naninfula Perch-Nielsen, 1968 emend Perch-Nielsen, 1971; Amitha Shafik, 1989; Notiocyrtolithus Shafik, 1989; Ommatolithus Shafik, 1989; Cruxia Varol, 1989 {multicyclic rhabdoliths with cap-shaped protrusions. Differential preservation and illustration modes (LM, TEM, SEM) makes it impossible to rationalise the taxonomy on current data}
• Rhabdolithus Kamptner ex Deflandre inGrass* 1952 {with hollow spine, without flaring collar. Shield includes outer cycle with approximately half the number of elements of the inner cycle. N.B. Many authors assign Eocene species to Rhabdosphaera but the structure of these forms appears significantly different to that of modern Rhabdosphaera}

Extant genera

Many species are too small to be readily identified by LM as isolated coccoliths and are rarely recorded as fossils. See Kleijne (1992) for review and detailed descriptions.

• Acanthoica* Lohmann, 1903 emend. Schiller, 1913 and Kleijne, 1992 {spines at poles only, coccoliths with radial T-cycle}
• Algirosphaera** Schlauder, 1945 emend. Norris, 1984 {spines modified into elongate domal or double-lipped (labiatiform) protrusion}
• Anacanthoica** Deflandre, 1952 {monomorphic, no spines, otherwise similar to Acanthoica}
• Cyrtosphaera* Kleijne, 1992 {vari-monomorphic, with domal or conical protrusions on all coccoliths, some species are strikingly similar to the Eocene genera, but with simple shields}
• Discosphaera* Haeckel, 1894 {monomorphic, spines trumpet-like (salpingiform)}
• Palusphaera** Lecal, 1965 emend. Norris, 1984 {monomorphic, long spines}
• Rhabdosphaera* Haeckel, 1894 (= RhabdolithusKamptner ex Deflandre in Grass*, 1952) {dimorphic with spine and non-spine bearing coccoliths, distributed around coccosphere. Used for many fossil rhabdoliths}

1.2. Placolith heterococcoliths

Order PRINSIALES Young & Bown 1997

Description: Extant species are based on non-motile heterococcolith-bearing stages. In addition, at least Emiliania huxleyi and Gephyrocapsa oceanica have a motile scale-bearing stage. Coccospheres are subspherical and monomorphic. Coccoliths are placoliths but unlike the Coccolithaceae, growth does not occur downward from the proto-coccolith ring. The R-unit is always well developed, forming a proximal shield-element, two tube-elements with opposite senses of imbrication, and usually a central-area element. The V-unit is well developed in early forms, the Toweius -type structure, forming an upper layer to the proximal shield, an outermost tube and the distal shield. In the Reticulofenestra -type structure, the V-unit is virtually absent and the outer of the two R-unit tube-cycles is extended to form the distal shield. In both structure types, the locus of the proto-coccolith ring is usually marked by a ring of slits. Central-area structures are always conjunct, being formed from either the central-area element or the inner tube-element of the proximal shield.

Comment: The major difference between the Reticulofenestra -type structure and the Toweius-type structure forms a useful basis for subdividing this group (see Young & Bown, above, Figure 1), which was previously considered of family level and has been variously referred to as the Noelaerhabdaceae and Prinsiaceae. Both family names are valid and so are used for the two emended families.

Family PRINSIACEAE Hay & Mohler, 1967 emend. (V-unit prominent)

Description: Genera with a prominent V-unit, and so a dark distal shield in LM (Toweius-type structure - see description of order).

Comments: Confined to the Paleogene. They can be difficult to separate from small Coccolithus species in the LM, despite the great structural differences; details of central-area structure and the extinction figure need to be used.

• Futyania Varol, 1989 {tube-elements extended to form a flower-like distal structure}
• Girgisia Varol, 1989 {modified Toweius-type structure - central-area open, proximal shield shows low birefringence and appears to be monocyclic but visible cycle must be an R-unit. Monospecific, G. gammation}
• ?Genus Hornibrookina Edwards, 1973 {narrowly elliptical placoliths with central-area filled by large bars; proximal shield monocyclic; distal shield bicyclic with inner cycle forming crown-like structure}
• Neobiscutum Varol, 1989 {earliest Danian forms, possibly with a simpler structure}
• Praeprinsius Varol & Jakubowski, 1989 {small early Prinsius/Neobiscutum intermediates}
• Prinsius Hay & Mohler, 1967 {elliptical, central-area closed by plate}
• Toweius Hay & Mohler, 1967 {circular to subcircular, central-area with variable number of pores}

Family NOELAERHABDACEAE Jerkovic, 1970 emend. (V-unit vestigial)

Description: Coccoliths with Reticulofenestra-type structure, i.e. V-unit vestigial, R-unit forms proximal shield, distal shield, inner and outer tube-cycles, grill and any central-area structures; strongly birefringent (see also description of order Young (1989)).

• Bekelithella Bona & Gal, 1985 {with flaring circlet of spines, formed from inner tube-elements, only recorded from Paratethys}
• ?Craterolithus Firth, 1988 {distal shield large and flaring upwards with 10-12 spines projecting down from it}
• Cribrocentrum Perch-Nielsen, 1971 {central-area partially closed by extensions of the inner tube-elements}
• Cyclicargolithus Bukry, 1971 {(sub)circular with narrow central opening, often regarded as a junior synonym of Reticulofenestra }
• Dictyococcites Black, 1967 {central-area closed by plates formed from inner tube-cycle. N.B. Paleogene species such as D. scissura appear distinct, but Neogene forms sometimes assigned to Dictyococcites are probably heavily calcified varieties of Reticulofenestra}
• Emiliania* Hay & Mohler in Hay et al ., 1967 {slits between all distal shield, and some proximal shield, elements}
• Gephyrocapsa* Kamptner, 1943 {with conjunct bar, formed from inner tube-elements}
• Noelaerhabdus Jerkovic, 1970 {with spine, formed from inner tube-elements, only recorded from Paratethys}
• Pseudoemiliania Gartner, 1969 {slits between some distal shield elements}
• Reticulofenestra* Hay, Mohler & Wade, 1966 (= Apertapetra Hay, Mohler & Wade, 1966) {lacking distinctive features, or with central-area partially closed by extensions of the inner tube-elements}
• [Bicolumnus Wei & Wise, 1990] {this morphotype is very similar to Pyrocyclus; it probably represents the isolated central-area and tube-cycles of Dictyococcites}
• [Crenalithus Roth, 1973] {often used for small Reticulofenestra species, but the holotype is a junior synonym of Gephyrocapsa oceanica}
• [Pyrocyclus Hay & Towe, 1962] {used for small 'species' with open central-area and no real shield development; these are probably early growth stages and broken specimens of other reticulofenestrids, Young (in press)}.

Order COCCOSPHAERALES Haeckel, 1894 emend.

Comments: Extant species form non-motile heterococcolith-bearing stages. In Coccolithus and Calcidiscus, these are known to alternate with motile holococcolith-bearing stages. The family Coccolithaceae is often used for all placoliths not placed in the Prinsiales. Nonetheless, typical members of the family have a very characteristic rim-structure whilst many other members have modified versions of this structure with sufficient similarity to strongly suggest a common origin. In particular, growth occurs downward from the proto-coccolith ring which consequently becomes embedded within the rim. Hence, on intact specimens, there is no obvious belt of alternating V- and R-elements, but such a belt is seen on specimens where the proximal shield has been partially detached.

Family COCCOLITHACEAE Poche, 1913 emend.

A. Genera with Coccolithus-type rim

Description: These have the Coccolithus-type rim-structure, as described in Young (1992). The V-unit forms both the distal shield and the proximal layer of the central-area (= centro-proximal cycle). The R-unit forms the proximal shield and the distal layer of the central-area (= centro-distal cycle). The proximal shield itself is bicyclic with distinct upper and lower layers but these are both formed from the R-unit, unlike the Toweius-type structure. The central-area is often spanned by disjunct structures and these are used to define genera.

• Bramletteius Gartner, 1969 {Cruciplacolithus-like base with very large monocrystalline 'paddle' or spine}
• Campylosphaera Kamptner, 1963 {strongly convex shield giving subrectangular outline, axial cross in central-area}
• Chiasmolithus Hay, Mohler & Wade, 1966 {diagonal, usually offset, cross in central-area, the bars of which show a median extinction line in XPL; the centro-distal cycle forms a distinct collar around contact with V-units}
• Clausicoccus Prins, 1979 {typical Coccolithaceae rim (see SEMs in Varol (1989)), central-area wide, filled by disjunct plate with variable number of perforations}
• Coccolithus* Schwartz, 1894 (= CoccosphaeraWallich, 1877; Ericsonia Black, 1964; CyclolithusKamptner, 1948) {Central-area open or with a disjunct transverse bar}
• Cruciplacolithus Hay & Mohler in Hay et al., 1967 {axial or near-axial cross in central-area}
• Sullivania Varol, 1992 {diagonal, usually offset, cross in central-area, bars undivided in XPL. Centro-distal cycle does not form a distinct collar}

B. Genera with ?Slightly modified Coccolithus-type rim

The following genera show strong similarities to the typical Coccolithaceae but probably have somewhat modified rims.

• [Birkelundia Perch-Nielsen, 1971] {Perch-Nielsen (1971) placed three Eocene species in this genus on grounds that they had a monocyclic proximal shield, however, none of them are unambiguous}
• Coronocyclus Hay, Mohler & Wade, 1966 {open ring-like coccolith without shields, elements of rim complexly intergrown, apparently with outer V-unit and inner R-unit. Included here in Coccolithaceae as structure suggests it is a neomorphic Coccolithus derivative}
• Crassidiscus Okada, 1990 {monospecific, C. backmanii , large form with indistinct XPL image; SEM shows 3 tiers which may be equivalent to the three shield cycles of Coccolithus }
• Hughesius Varol, 1989 {like Clausicoccus but no inner bright cycle, so centro-distal R-unit element probably missing; ?proximal shield also formed from V-unit}
• Solidopons Theodoridis, 1984 {narrow rimmed ?placolith with prominent arched bridge; extinction figure suggests affinities to Coccolithus}

Family CALCIDISCACEAE Young & Bown 1997. ( Calcidiscus-type rim)

Description: Dominant phase of life-cycle, non-motile with placolith heterococcoliths. V-unit forms the distal shield and tube, extending to the proximal surface. R-unit forms the proximal shield. As in the Coccolithaceae, growth occurs downward from the proto-coccolith ring which becomes embedded within the structure so that alternating V- and R-units are only visible on specimens where the proximal shield has broken off. Distal shield sutures typically show laevogyral curvature. The proximal shield is usually monocyclic with radial sutures; sometimes it is bicyclic due to the development of a lower layer, with elements showing strong dextral obliquity (in proximal view). The connection between the proximal and distal shields is weak and they frequently separate.

Comments: These genera have previously been included within the Coccolithaceae but the distinctively different structure appears to warrant classification in a separate family. The cytology of Umbilicosphaera is described by Inouye & Pienaar (1984).

• Calcidiscus* Kamptner, 1950 (= Cyclococcolithus , Cyclococcolithina, Cycloplacolithella, Cycloplacolithus, Tiarolithus , Striatococcolithus Bukry, 1971) {(sub)circular, central-area closed or narrow. Proximal shield elements often kinked, sometimes becoming bicyclic}
• Cryptococcolithus Gartner, 1992 {elliptical, proximal shield thin so coccolith is dark in XPL, central-area with non-birefringent perforate plate}
• Cycloperfolithus Lehotayova & Priewalder, 1978 {subcircular, central-area with non-birefringent perforate plate. Often regarded as a junior synonym of Calcidiscus but proximal shield is described as bicyclic}
• Geminilithella Backman, 1980 {wide central-area and narrow rim (see Young (in press)) better regarded as a junior synonym of Umbilicosphaera}
• Hayaster* Bukry, 1973 {proximal shield diminutive, distal shield with straight radial sutures and angular ray ends, 9-13 elements}
• Oolithotus* Reinhardt in Cohen & Reinhardt, 1968 {asymmetrical, proximal shield elements show complex kinking, nearly becoming bicyclic, otherwise very like Calcidiscus}
• Umbilicosphaera* Lohmann, 1902 {open central-area, distal shield elements show complex kinked sutures. Proximal shield monocyclic or bicyclic}
• [Striatococcolithus Bukry, 1971] {the only species, S. pacificanus, should probably be included in Calcidiscus }

Family PLEUROCHRYSIDACEAE Fresnel & Billard, 1991

This family is here included in the Coccosphaerales since the rim structure appears to be a simplified version of that of the Coccolithaceae. It is monogeneric and possibly should be subsumed into the Coccolithaceae.

• Pleurochrysis** Pringsheim, 1955 (= Cricosphaera Braarud, 1960) {Coccolithophore motile, neritic, coccosphere monomorphic. Coccoliths are narrow-shielded placoliths. Tightly interlocked crystal-units, V-unit forms distal shield and tube. R-unit forms proximal shield and small element on inside of tube - cricoliths}

Order WATZNAUERIALES Bown, 1987

Family WATZNAUERIACEAE Rood, Hay & Barnard, 1971

• Cyclagelosphaera Noel, 1965 {this predominantly Mesozoic genus persisted into the Danian; see Bown & Young (above)}

1.3. Heterococcoliths of uncertain affinities

Family HYMENOMONADACEAE Senn, 1900

Description: Small littoral and fresh-water coccolithophores. Coccoliths are goblet-shaped muroliths with open central-area, well-developed proximal flange, and a narrow distal flange or flaring end, entirely formed of a single cycle of <15 crystal-units (tremaliths). In Ochrosphaera, crystal-units have subvertical orientations. They differ from Pleurochrysidaceae by the absence of a second cycle of units (i.e. R-units). References include Manton & Peterfi (1969), Braarud (1954) and Fresnel (1994).

• Hymenomonas** Stein, 1878 {freshwater and marine species, coccoliths with distal part flaring, elements have pointed ends}
• Ochrosphaera** Schussnig, 1930 {littoral, coccoliths with distinct distal flange}

Family PAPPOSPHAERACEAE Jordan & Young, 1990

Description: Family of minute, lightly-calcified coccolithophores, mainly known from high-latitudes, with holo- and heterococcolith phases (Thomsen et al., 1991). Heterococcoliths have a narrow murolith rim; +/- open central-area; tall, delicate spine supporting calyx of four plates (pappoliths). Holococcoliths tower-like, crystallites arranged in hexagonal or triangular groups.

A. Genera definitely related to Papposphaera

• Pappomonas** Manton & Oates, 1975 {heterococcospheres dimorphic, only circum-flagellar coccoliths have spines}
• Papposphaera** Tangen, 1972 {heterococcospheres monomorphic, all bear spines}
• Trigonapsis** Thomsen, 1980 {tower-like holococcoliths with triangular crystallite groups}
• [Turrisphaera** Manton, Sutherland & Oates, 1976] {tower-like holococcoliths with hexagonal crystallite groups, = holococcolith phases of Pappomonas and Papposphaera spp.}

B. Possibly related weakly-calcified holococcoliths

• Balaniger ** Thomson & Oates, 1978 {coccoliths are organic scales with a few pyramidal ?crystallites}
• Calciarcus** Manton, Sutherland & Oates, 1977 {rhombohedral crystallites forming 4 struts ?with calcareous rim}
• Quaternariella** Thomsen, 1980 {coccoliths are organic scales with a few rhombohedral crystallites}

C. Possibly related weakly-calcified heterococcoliths

• Jomonlithus** Inouye & Chihara, 1983 {murolith coccoliths with Wigwamma-like rim, no central-area structures; partially calcified specimens show beaded ultrastructure}
• Wigwamma** Manton, Sutherland & Oates, 1977 {simple rim and 'wigwam' of 3 or 4 struts}

Cenozoic heterococolith genera incertae sedis

A.Genera with a fossil record

This is a diverse group but all show typical heterococcolith structure, only Umbellosphaera and Neosphaera are extant.

• Calciopilleus Muller, 1974 {bell-shaped with external ridges}
• Conococcolithus Hay & Mohler, 1967 {poorly-documented conical placolith, one species, C. minutus, Paleocene}
• Ellipsolithus Sullivan, 1964 {placolith morphology but structure anomalous}
• Hayella Gartner, 1969 (= Nannocorbis Muller, 1974, cf. Theodoridis, 1984) {tube with two flanges, ?a modified placolith, formed of a single cycle of subvertical crystal-units}
• Ilselithina Stradner in Stradner & Adamiker, 1966 {modified placolith, distal shield reduced to a cycle of spines, formed of single cycle of steeply inclined crystal-units}
• Markalius Bramlette & Martini, 1964 {moderately birefringent interference figure with a bright tube-cycle; central-area narrow or closed - details of structure uncertain}
• Neosphaera* Lecal-Schlauder, 1950 (= Craspedolithus Kamptner, 1963) {open ring-like coccolith with proximal shield only, formed of single cycle of subvertical crystal- units; possibly an alternate life-cycle stage of Ceratolithus (Alcober & Jordan, 1997)}
• Pedinocyclus Bukry & Bramlette, 1971 (nom subst pro Leptodiscus Bukry & Bramlette, 1969) {poorly-documented circular placolith showing low birefringence}
• Tintinnabuliformis Varol, 1991 {bell-shaped with apical horns}
• Umbellosphaera* Paasche in Markali & Paasche, 1955 (= Ellipsodiscoaster Boudreaux & Hay, 1969) {? motile, placolith-like morphology with distal shield greatly extended, R-unit forms central-area, tube and distal shield. Diminutive V-unit forms very narrow proximal shield. Distal shield is thin, except in some U. tenuis, and so shows low birefringence; tube highly birefringent}

B. Genera only known from the plankton

This diverse group of genera all show basic heterococcolith features (except perhaps Polycrater) and are mostly small and poorly known. (N.B. We place Florisphaera among the nannoliths, rather than in this group, since it lacks any basal disc structure.)

• Gladiolithus* Jordan & Chamberlain, 1993 {basal disk of two elements supporting a long hexagonal-section spine; in LM isolated spine fragments resemble long-thin Florisphaera profunda coccoliths}
• Polycrater** Manton & Oates, 1980 {coccoliths are aragonitic square-section cones, c.1µm across, very numerous on coccosphere}
• Turrilithus** Jordan et al., 1991 {narrow-rimmed placolith base, square-section flaring spine}
• Vexillarius** Jordan & Chamberlain, 1993 {small and rare, murolith base, square-section flaring spine}
• [Thorosphaera** Ostenfeld, 1910] {very poorly-documented large coccolithophore, with tube-like coccoliths, possibly Scyphosphaera . N.B. Thorosphaera flabellata is now placed in Gladiolithus}

Return to: top; Introduction, family-level overview, Mesozoic heterococcoliths, holococcoliths & nannoliths; Cenozoic heterococcoliths, holococcoliths & nannoliths

This page was produced by Jeremy Young, feedback and corrections welcome.