Neogene vs. Paleogene discoasters: Although discoasters were abundant and diverse during most of the Paleogene only a single species, D. deflandrei, survived in the Late Oligocene and all Neogene discoasters evolved from it (Prins 1971). So, they form an homogeneous group, and they are separated from typical Paleogene species by a range of characters (N.B. a number of Paleogene species, e.g. D. okadai, D. septemradiatus, more closely resemble the Neogene group than typical Paleogene discoasters).
The most important of these characters are:
A. Overall shape: Palaeogene discoasters typically have their rays in contact for most of their length giving a ""rosette"" shape, in contrast to ""star"" shaped Neogene discoasters (Aubry 1984).
B. Ray shape: The rays of Paleogene discoasters are often curved and asymmetrical, whereas Neogene discoasters nearly always have straight bilaterally symmetrical rays.
C. Number of rays: Paleogene species usually have more rays (8-30) than Neogene ones (5 or 6, rarely 3-8).
D. Ray attachment surface: The rays of Paleogene species usually join along inclined and curved surfaces, whereas the attachment surfaces of Neogene species are planar and vertical (Theodoridis 1984).
It would be reasonable to treat the two groups as separate genera, but there are complications in doing this, due to the fact that Tan (1927, 1931), who first described discoasters, coined a number of generic names without regard to nomenclatural legalities. Theodoridis (1983, 1984) argued that the name Discoaster was invalid and that instead the names Helio-discoaster and Eu-discoaster should be used for, respectively, the Paleogene and Neogene groups. The validity of his argument has not been accepted, and almost all workers have continued to use the name Discoaster in the traditional sense.

Crystallography: Black (1972) analysed the crystal faces developed during overgrowth of Neogene discoasters. He showed that the rays have radially symmetrical crystallographic orientations, and that, as a result of the low symmetry of calcite, the two faces are crystallographically distinct. The effect of this is most consistent in the central area; on one side two crystal faces are readily developed resulting in a radial ridge. On the other side only one face is preferentially developed, leading to radial flats. Black (1972) termed these respectively the E- and F- surfaces. However they correspond to the proximal and distal surfaces, and the central area structures to radial knobs and depressions.

Neogene-discoaster morphology: The ancestral Neogene discoaster species, D. deflandrei has a rather simple form; the two sides are similar, and lack elaborate central area structures. Subsequent species show increased complexity, with development of a range of structures. There is, however, a pattern to these structures. In particular different structures occur consistently on the two surfaces. It is convenient to differentiate these two surfaces as proximal and distal. It is, of course, not known how or in what orientation discoasters where born on the nannoplankton cell. However, it seems reasonable to take the analogy of coccoliths and assume that for concavo-convex species the concave side was innermost. On this basis the concave side can be termed proximal, and the convex side distal, as recommended by Farinacci (1971) and Young et al. (1987).
The various structures developed are illustrated in the figure, using D. surculus as an example, since it shows the greatest range of features. On the distal side there is a distinct central area formed by the rays widening and uniting. Within this central area the rays may be slightly depressed and/or separated by low sutural ridges. In the middle of the central area there is often a stellate knob the arms of which point toward the ray sutures. Away from the central area the rays extend nearly horizontally, with a rather flat distal surface, on which distal ridges may occur. These distal ridges are always confined to the rays, never running into the centre of the discoaster. Bifurcations occur at the tips of rays in many species, these are primarily formed from the distal surface of the ray.On the proximal side there is no clear central area / ray division, instead ridges run continuously from the ray into a central knob. This proximal knob thus has a radial stellate form, in contrast to the inter-radial distal knob. The proximal ridges may run continuously along the rays and then build downwards, giving the discoaster a concavo-convex form. The contrast between proximal ridge and distal surface gives the rays an asymmetrical section.
Since rays are formed of single crystals this division of the rays into separate structural elements is essentially artificial. Nonetheless they are recognisable and used together such features as convexity, central knob orientation, and central area development allow consistent differentiation of proximal and distal surfaces. Some species show much stronger development of the features of one side or the other. For instance D. brouweri has a well developed proximal side: it has little or no central area, distal knob, or bifurcations; but the proximal blades, and proximal ridges are well developed, and there is often a proximal knob. D. deflandrei by contrast has a virtually featureless proximal surface but well developed distal features (particularly central area and bifurcations).

References: