The presence of radiolarian taxa, together with general information on coarse-fraction composition, are given in Table T1. Almost all slides contained common to abundant well-preserved or moderately well preserved radiolarians. Both pennate and centric diatoms were common to abundant in many slides but were rarely so abundant as to significantly obscure or dilute the radiolarian faunas. Silicoflagellates, sponge spicules, and lithic fragments (both ash and rock fragments) were more sporadically present, although they were common in some intervals in the section. No obvious relationship between the presence of these latter components and other aspects of the assemblages (abundance and preservation of radiolarians and diatoms) could be detected.
Radiolarian faunas from Site 1138 are all of a typical Neogene Antarctic biogeographic affinity, with robust, high optical contrast shells and a dominance of prunoids (Pylonidae), Actinommidae, and Plagiacanthidae, particularly members of the genus Antarctissa (Lazarus and Caulet, 1994). No subantarctic faunas were observed. The standard stratigraphic marker species (Lazarus, 1992) for this time interval were all seen and in most samples were easily identifiable. In addition to the standard marker species, we noted the distribution of several other common taxa that are typical of late Miocene to early Pliocene Antarctic radiolarian faunas and include as well one species—Cenosphaera sp. Keany (1979)—whose presence has until now not been reported in Deep Sea Drilling Project (DSDP) or ODP sections, but whose distribution may be of stratigraphic value. Except as noted below, all taxonomic concepts used are those given in Lazarus (1990, 1992). Some difficulties were noted with certain taxa in determining the morphologic limits to closely related species, which make the precise placement of some of the first appearance datums (FADs) and last appearance datums (LADs) difficult.
Also, in many slides at least some specimens were seen that partly fit a species description but that have characteristics that exclude them from being positively identified. Such specimens are marked on a chart by a question mark (?) or by "sl" (sensu latu). Given these occasional taxonomic and preservational problems, an empty cell in the chart indicates that the nominate taxa was not seen, but not that it was necessarily absent. Specific slides were searched to confirm the absence of marker taxa; such data cells are marked with a dash.
Cenosphaera sp. Keany 1979 (Plate P1, figs. 4-13).
Remarks: although only fragments of this form have been seen in our study, the distinctive short forked radial spines on the outside of the cortical shell and the moderately large, irregular pores strongly resemble a form illustrated by Keany (1979), which he referred to as Cenosphaera sp. Chen (Chen, 1975). Chen's (1975) illustration, however, does not appear to have either the forked spines or irregular pores of Keany's form, and we believe the two are not conspecific. Cenosphaera sp. Keany 1979 appears to have a restricted stratigraphic range in the Antarctic, although further work is needed to ensure that this species can be reliably distinguished from Heliosomma watkinsi Keany 1979, even in fragmentary specimens. The latest Miocene (basal Gilbert magnetochron) absolute age range assigned to this species by Keany is questionable, as it was found by him only in a single piston core (E34-19) without any other positive biostratigraphic or paleomagnetic evidence that would allow an unambiguous absolute assignment of the section to the geochronologic timescale.
Prunopyle titan (Plate P2, fig. 18).
Remarks: specimens rounder than the late Miocene-early Pliocene form of P. titan used in Southern Ocean stratigraphy (Lazarus, 1990, 1992) were identified in many slides and are marked in the range chart with a question mark, as it is unclear if they really are conspecific with the marker taxon. It is also unclear if the late Miocene form of P. titan is really conspecific with the type specimens of the species, which were described by Campell and Clark (1944) from older North Pacific material.
Lithomelissa stigi (Plate P3, figs. 8-14).
Remarks: this species' range is discontinous, and it is morphologically very variable. A true L. stigi should be small, with large pores and a strong spine to one side. It is likely that more than one species has been included in our study in this category—many specimens identified as L. stigi were larger or smaller than usual, had small pores, or had only a very small spine.
Eucyrtidium cienkowski and Eucyrtidium calvertense (Plate P2, figs. 1-8).
Remarks: transitional forms are common in the lower part of the studied section where both the nominate species are present. Farther upsection, specimens that closely resemble E. cienkowski but have more ridged walls are found alongside typical E. calvertense specimens.
Triceraspyris coronata and Triceraspyris antarctica (Plate P2, figs. 10-16)
Remarks: although T. coronata was used as a formal stratigraphic marker by Weaver (1983), we have had great difficulty distinguishing this species reliably both from forms similar to T. antarctica and to several other forms that differ from typical T. coronata in the size of the pores, robustness of the shell, and several other characteristics. Only a very few of our specimens are as robust as those illustrated by Weaver (1983). A revision of the taxonomy of these forms may be needed in order to improve the reliability of T. coronata as a stratigraphic marker. See also discussion in Lazarus (1990, p. 716, 1992, p.797).
Cycladophora bicornis and Cycladophora pliocenica (Plate P3, fig. 7).
Remarks: transitional forms between C. bicornis and C. pliocenica are fairly common in some slides, although C. pliocenica sensu stricto is clearly absent in Sample 183-1138A-13R-2, 20 cm, and below and is present in Sample 183-1138A-12R-3, 20 cm.
Cycladophora spongothorax (Plate P3, fig. 16)
Remarks: specimens near the top and bottom of the range frequently have only traces of external sponge. However, they still possess a steplike external profile and externally developed thoracic rings, and thus are considered to belong to this taxon (Lombari and Lazarus, 1988).