Of the five holes drilled at Site 1201, Holes 1201B, 1201C, and 1201D recovered a nearly continuous sedimentary succession. Smear slides were made from core catchers and selected intervals of these three holes for nannofossil examination. Because one major objective of this site was to date the sediments contacting the basement, sediments that were situated immediately above the basalt and those sandwiched occasionally within the basalt were also examined.
Nannofossils are scarce in most of the core catchers we examined. Extra samples were taken from various carbonate-bearing intervals detected by HCl tests of the sediments. Moderately to well-preserved nannofossil assemblages were found in the bioturbated pelagic clays and mudstones intercalated within the turbidite sequence (see "Lithostratigraphy") (Table T5; Fig. F48). The fossil assemblages indicate that an expanded section of late Eocene-middle Oligocene age was recovered (Fig. F49). A detailed zonation of the sedimentary succession was difficult to prepare, partly because of the sporadic nature of the fossil occurrences and partly because of the inevitable reworking of old nannofossils into younger sections via turbidity currents. The frequent reworking makes the last occurrence (LO) of species less reliable for age determination. Unfortunately, the zonation scheme of the late Eocene-middle Oligocene sequence relies conventionally on many LOs of species.
To prepare a sensible zonation of the cored succession, special caution was taken to evaluate each of the first occurrences (FOs) and LOs of the age-diagnostic species. In the following section, we will first lay out our general consideration and weighing of the various datum levels in our record before giving the detailed description of the individual holes.
Almost every single nannofossil-bearing sample in the studied section contains this easily identified, morphologically distinctive species (Table T5; Fig. F49). Dictyococcites bisectus has a long range, between 38.0 and 23.9 Ma (Berggren et al., 1995). Both the FO and LO are considered reliable age markers (Wei, 1992; Berggren et al., 1995). The persistent presence of this species in our record ensures that the sequence belongs to the late Eocene-Oligocene.
In calcareous nannofossil biostratigraphy, the Oligocene/Eocene boundary (33.6 Ma) is conventionally approximated by the extinction of Discoaster barbadiensis and Discoaster saipanensis. The LO of D. saipanesis has been preferred as a more reliable marker near the middle of Subchron C13r at 34.2 Ma (Wei and Wise, 1989; Berggren et al., 1995). The LOs of both species were designated in Sample 195-1201D-40R-6, 10 cm (462.36 mbsf). However, in the entire 510-m-thick sedimentary succession, there are only two samples near 462.4 mbsf containing good quantities of both species (Table T5; Fig. F49). The designation of the LOs and thus the boundary between Zones NP20 and NP21 at this level is circumstantial. Nevertheless, the common occurrence of both species in two consecutive samples, 195-1201D-40R-6, 10 cm (462.36 mbsf), and 40R-6, 25 cm (462.51 mbsf) ensures that the upper Eocene occurs in our record. The lack of Isthmolithus recurvus appears to indicate that the interval is still younger than 34.8 Ma (Berggren et al., 1995).
It is noticeable that a few reworked specimens of D. barbadiensis were spotted in various samples in the upper sections (Fig. F49; Table T5).
The LO of Ericsonia formosa has been widely used to define the Zone NP21/NP22 boundary. This datum level, near the bottom of Chron C12r at 32.8 Ma (Berggren et al., 1995), has been considered geologically synchronous and an excellent marker in the low and mid-latitudes (Wei and Wise, 1989). The presence of E. formosa at Site 1201 has not been abundant or persistent (Fig. F49; Table T5). Judging from the absence of this otherwise optically distinctive species in the long interval between Cores 195-1201D-1R and 30R, we anchored its last occurrence at 31R-CC (378.19 mbsf) (Fig. F49).
The LO of Reticulofenestra umbilicus has been widely used to mark the upper boundary of Zone NP22 in the upper Oligocene (Martini, 1971). Wei and Wise (1989) determined that the datum is located two-thirds of the way down in Chron C12, at ~32.3 Ma (Berggren et al., 1995). To ensure taxonomic consistency with previous work, we followed the commonly accepted cutoff size of 14 µm (Backman and Hermelin, 1986) to separate this species from the smaller but similar form termed Reticulofenestra samodurovii (= Reticulofenestra dictyoda or Reticulofenestra coenura, according to Wei and Wise, 1989). R. umbilicus is common to abundant in the lower sections of Hole 1201D, with its LO designated at 261.96 mbsf (Sample 195-1201D-19R-CC, 13-19 cm) (Fig. F49).
The Spenolithus predistentus-Sphenolithus ciperoensis lineage evolved through the late Eocene and Oligocene. The evolution of this lineage provides three important datum levels for the nannofossil zonations of Martini (1971) and Okada and Bukry (1980). The first appearance of descendant species in such an evolutionary series has a great advantage over other age markers in assigning ages for the turbidite succession under study. However, the gradual evolution of this lineage hampers a consistent differentiation of the successive species by workers in the field when subjectivity gets involved in taxonomy (see discussions in Roth et al., 1971a, 1971b; Haq, 1972; Lang and Watkins, 1984; Moran and Watkins, 1988; Wei and Wise, 1989). Moreover, as noted by Wei and Wise (1989), the FO of Sphenolithus distentus in many localities is well below the LOs of R. umbilicus and E. formosa and is therefore in conflict with the original zonation scheme of Okada and Bukry (1980).
To circumvent the "Sphenolithus problem," we have adopted the criteria suggested by Moran and Watkins (1988) to differentiate species. They found that the most accurate characteristics for separating S. ciperoensis from S. distentus are the "extinction" lines of the proximal shield (basal disc of spines) under cross-polarized light using a light microscope and the width of the proximal shield (as originally proposed by Bramlett and Wilcoxon, 1967; Roth, 1970). Specifically, the descendant species, S. ciperoensis, is characterized by the following: (1) the proximal shield is wider than any point on the apical spine and (2) the "extinction" lines of the proximal shield are "chevronlike" and do not cross each other when the apical spine is at 45° to either polarizer. In contrast, the proximal shield of S. distentus is rarely wider than the base of the apical spine and the "extinction" lines are V-shaped and crossed (Moran and Watkins, 1988). Similarly, we differentiated S. predistentus from S. distentus by the criteria stated in Roth (1971b): "If the angle between the extinction line and the median axis is 90° or less, (the) forms are assigned to S. predistentus" (p. 1106 of Roth, 1971b).
At Site 1201, S. predistentus is the most common and long-ranged species within the lineage, rendering it less useful in subdividing the sequence into zones. The marker of the top boundary of Zone NP24 (27.5 Ma), the LO of S. distentus, was found at 71.78 mbsf (Fig. F50) (Section 195-1201B-10X-1, 68 cm), together with the last appearances of several other species, including Bramletteius serraculoides, Discoaster tani ornatus, S. pseudoradians, and Helicosphaera compacta (Fig. F50). The truncation of so many species simultaneously appears to suggest the existence of a hiatus.
The FO of S. ciperoensis, the marker of the base of Zone NP24, was observed at 72.98 mbsf, only slightly below the LO of S. distentus (the top of Zone NP24) (Fig. F49). Zone NP24 in Hole 1201B, defined accordingly, is therefore extremely short. In view of the shortness of Zone NP24, as well as the truncation pattern of many concurrent species, we interpret that a hiatus exists at about this level and that Zone NP24 is truncated. The truncation of ranges at 71.78 mbsf, however, is an artifact caused by the poor recovery of Cores 195-1201B-10X and 9X (Fig. F49; also see "Lithostratigraphy"). The inferred hiatus, instead, is placed at 53.4 mbsf in Section 195-1201B-8X-4, where a distinct lithologic and color change separates lithologic Unit I from the underlying Unit II (see "Lithostratigraphy").
Figure F50 summarizes our interpretation of the ranges of these age-diagnostic species as well as several additional species. The range chart shows clearly that six NP zones were recognized throughout the sedimentary succession.
Nannofossils are generally scarce within the 90 m of Hole 1201B. Most core catchers are barren of nannofossils, except for a few samples in which traces or only a few specimens were spotted. Careful inspection and extensive sampling through the cored sections eventually yielded a few samples that contained sufficient nannofossils for biostratigraphic investigation (Figs. F48, F49). Section 195-1201B-4H-3 is exceptional in that it contains abundant, diversified, but only moderately preserved assemblages (Table T5). The consistent occurrence of Dictyococcites bisectus in Hole 1201B constrains the succession to be older than 23.9 Ma (late Oligocene). Indeed, the assemblages are dominated by Cyclicargolithus floridanus, with common occurrences of Sphenolithus moriformis and Discoaster deflandrei, typical of Oligocene tropical flora. Sample 195-1201B-4H-3, 68 cm (29.88 mbsf), witnessed the LO of S. ciperoensis, indicating an age older than 24.5 Ma (Berggren et al., 1995) (see also "Biostratigraphy" in the "Explanatory Notes" chapter). The lingering occurrence of S. ciperoensis continues downsection to Sample 195-1201B-10X-CC, 38-43 cm (72.98 mbsf), with the coexistence of S. distentus. The concurrent range of these two age-diagnostic species brackets an age of 27.5-29.9 Ma in Zone NP24 for the interval 71.8-73 mbsf. The shortness of Zone NP24 and the simultaneous truncation pattern of species, together with the distinct lithologic change at 53.4 mbsf, collectively suggest the existence of a minor hiatus. We infer that the upper part of Zone NP24 is missing (Figs. F49, F50).
Hole 1201C is barren of nannofossils except for a few intervals that show traces of coccoliths (Table T5). Although lithology and various physical properties (e.g., magnetic susceptibility, color reflectance, and velocity) indicate that Hole 1201C is almost identical to Cores 195-1201B-1H to 5H, the nannofossil-bearing intervals in Section 195-1201B-4H-3 do not have corresponding counterparts in Hole 1201C. Such a discrepancy indicates that nannofossils were favorably preserved in certain localized pockets or patches and that these pockets were not laterally continuous.
Because of the high sedimentation rates and quick burial associated with turbidites, the long sedimentary succession in Hole 1201D consistently yields moderately to poorly preserved nannofossils. Except for the successive appearances of D. saipanensis, E. formosa, and R. umbilicus, the assemblages are almost monotonous throughout the succession. Variation in the nannofossil composition, if any, is dictated by dissolution intensity (Fig. F48). Despite the nuisance caused by dissolution and reworking, the nannofossil record allowed us to subdivide the sequence into four zones: NP19/NP20, NP21, NP22, and NP23 (Figs. F49, F50). Most of Hole 1201D (80 to ~425 mbsf) belongs to the lower Oligocene, but the part of the succession below 462 mbsf is definitely older than 34.3 Ma (Table T6).
The lack of any nannofossils in the lowermost section (195-1201D-40R-7 to 53R [463-523 mbsf]) makes it impossible to gauge an age for the basement. In middle Eocene-early Oligocene time, Site 1201 would have been located in the equatorial zone (using the paleogeographic reconstruction of Cambray et al., 1995). The occurrence of abundant sphenoliths and discoasters in the succession and the lack of any high-latitude dwellers such as Chiasmolithus is consistent with this paleogeographic reconstruction. The lack of any calcareous microfossils in the lower section also fits the history of the calcite compensation depth (CCD) in the equatorial Pacific. Van Andel et al. (1975) documented clearly that the CCD dropped during the Eocene-Oligocene transition. It is likely that the shallow CCD during the Eocene prevented calcareous microfossils from being preserved. On the other hand, we have observed Eocene nannofossils from Zone NP19/NP20 for the first time on the sedimentary apron of the western flank of the Palau-Kyushu Ridge. In comparison, the oldest sediments above the basalt contact drilled at DSDP Site 447 are middle Oligocene (NP23) in age (Martini, 1981), whereas at Site 290 they are latest Eocene in age (NP20) (Ellis, 1975).
Section 195-1201D-44R-4, 140 cm (499.6 mbsf), contains radiolarians in red clay (or porcellanite). Potential utility of the radiolarians for biostratigraphic age determination at this site awaits further exploration. Despite extensive sampling and close examination of sediments sandwiched in the basalt section in Cores 195-1201D-45R to 53R, no nannofossils were found. Instead, examination of thin sections of interpillow fillings in the basalt sections revealed traces of microfossils, possibly radiolarians (e.g., see Fig. F31B for a biogenic clast found in the interpillow sediments). These biotic remains possess great potential for revealing the ages of these sediments.
At Site 1201, a 509-m-thick sedimentary succession was recovered above the basaltic basement. The topmost (0-26.5 mbsf) and lowermost (462-509 mbsf) sections are barren of nannofossils. Although traces of diatoms and ichthyoliths were spotted in the top section, the lack of calcareous nannofossils in the upper 26 m makes it impossible to assign any ages for the top section. The upper pelagic clay of Unit I is thinner than similar units cored at DSDP Sites 290 (~90 m) and 477 (37.5 m) located south of Site 1201. The uppermost unit at DSDP Site 290 was thought to be early Pliocene, underlain by Oligocene nannofossil-bearing clay, whereas the top unit in DSDP Hole 477 was regarded to be early Miocene in age based on fossil fish teeth. It is premature to correlate the top section (0-26.5 mbsf) of Site 1201 to either of the DSDP sites or to draw any conclusion about the nature of the hiatus based on the shipboard biostratigraphic investigation. The lack of post-Oligocene-age calcareous microfossils in the top section of Site 1201 has been attributed to the submergence of the ocean floor through the CCD after the early late Oligocene (Scott et al., 1980).
Moderately to poorly preserved nannofossils in the turbidite succession (lithostratigraphic Unit II) allowed us to recognize six biozones spanning from Zone NP19/NP20 to NP25 (Fig. F50). The estimated ages of the major bioevents are listed in Table T6. The turbidites between 54 and 462 mbsf represent an expanded sequence of late Eocene-early Oligocene age consisting of the top of Zone NP19/NP20, Zones NP21, NP22, NP23, and a truncated Zone NP24. Separated by a short hiatus, lying on top of the turbidites, is a 25-meter-thick sequence of upper Oligocene (29.4-54.3 mbsf) (NP25) red claystone (Fig. F50). The age of the basal 47 m of sediments above the basement contact is unknown. Nevertheless, compared to the previous drilling results from DSDP Sites 290 and 477, the Eocene sediments (>34.3 Ma) recovered at this site are the oldest observed so far on the sedimentary apron of the Palau-Kyushu Ridge. Radiolarians found in Section 195-1201D-44R-4, 140 cm (499.6 mbsf), may have potential for constraining the age of the bottom sediments.
Overall, the recovered succession at Site 1201 is similar to the material recovered from DSDP Sites 290 and 447 located south of this site. However, Site 1201 is unique in having an expanded early Oligocene section, which provides a detailed high-resolution record of the history of the Palau-Kyushu Ridge.