Planktonic foraminifera from the Pliocene Coralline Crag of Suffolk, Eastern England

Nine species of planktonic foraminifera have recently been identified from the Coralline Crag of Suffolk, Eastern England. Stratigraphically important species include Globorotalia puncticulata (Deshayes) and Neogloboquadrina atlantica (Berggren) whose joint ranges, based on evidence from the North Atlantic, indicate an age for the deposit of between c. 4.2 and 2.3 Ma; other published evidence suggests that it may not be younger than c. 3.4 Ma. The low planktonic foraminiferal species diversity is interpreted as being due to a combination of factors, including the shallow water nature of the Coralline Crag, which would preclude deeper water species of planktonic foraminifera, and the possibility that the source of the fauna was from relatively high latitudes of the North Atlantic via the northern entrance to the North Sea. The planktonic foraminifera recorded indicate that the water temperature of the sea was within the range 10–18°C. The Coralline Crag contains very rich Lusitanian faunas of gastropod and bivalve molluscs, bryozoans, ostracods and benthic foraminifera.


INTRODUCTION
Foraminifera from the Coralline Crag, a richly fossiliferous shallow marine deposit, which adjoins and underlies part of the southern North Sea (Fig. 1). were first recorded and systematically illustrated by Jones et al. (1866). At that time the only indigenous planktonic specimens were recorded as Globigerina bulloides d' Orbigny . A number of slides in the foraminiferal collections of the British Museum (Natural History) contain planktonic foraminifera from the Coralline Crag: H. B. Brady (P 38449;Sutton Knoll), H. B. Brady (P 38436; Sudbourne Park), C. D. Ovey (ex P 38547; Sudbourne Park), F. Mockler (P 16385, P 16387, P 16388; Gedgrave Cliff), W. K . Parker & T. R. Jones (P 48315. P 48356; Sutton Knoll) (for localities, see Fig. 1; for occurrences, see Table 1). All these slides are labelled "Globigerina bulloides d'orbigny", but in fact contain exclusively Neogloboquadrina atlantica (Berggren) .
which is now found to be the commonest species in the Coralline Crag. The actual specimen illustated by Jones et al. (1866, pl. 2, figs. 1, 2) cannot be located with certainty, but it is probably one of a suite of specimens from Sutton Knoll in the Parker & Jones Collection; it is undoubtedly referable to N . atlantica, and in possessing a high spire, is similar to that illustrated by us in PI. 2, figs. 5-7. Subsequent work by Carter (1951;1957). Wilkinson (1980) and Doppert (1985) have concentrated almost solely on the benthic foraminifera. The most recent work is by Hodgson & Funnel1 (1986) who have re-interpreted the benthic foraminifera, particularly from a palaeoecological viewpoint, and some reference is made in their paper to the quantitative occurrence of planktonic species formally described herein. King (1983) recorded and illustrated the following species of planktonic foraminifera from the Pliocene of the North Sea: Globorotalia inflata (d'orbigny), G . puncticulata (Deshayes) , Neogloboquadrina atlantica (Berggren) and N . pachyderrna (Ehrenberg); also N . atlantica from the Coralline Crag. From the evidence of the benthic foraminifera King (1983, 14) suggested that the Coralline Crag was coeval with his (benthic) Early Pliocene Zone NSB 14 and from the presence of sinistrally coiled N . atlantica he indicated a correlation with his (planktonic) Late Miocene-Early Pliocene Zone NSP 15 (see Table 2). This paper describes and illustrates the planktonic foraminifera obtained from samples collected by D . Curry at Sutton Knoll and Gedgrave Cliff and by N. J . Morris at Ramsholt Cliff. See Fig. 1 (Mathers et al., 1984). This represents a lower energy environment than much of the Coralline Crag, and molluscs and bryozoans from it are well preserved. This facies probably corresponds to the Lower Division (15m) (Prestwich. 1871) of the total (25m) thickness of the Coralline Crag (see Mathers et a [., ). Doppert (1985 has suggested that the benthic foraminifera from these localities may indicate slightly older deposits than the remainder of the Coralline Crag, corresponding to the lower part, rather than the main part of the Netherlands FB Zone. This is consistent with the correlation of the Coralline Crag with the Early Pliocene NSB 14 Zone of King (1983). Andrew & West (1977) equated the samples from the Orford Borehole (depth 33-38 feet), on palynological evidence to the Brunssumian (Early Pliocene) Stage of the Netherlands succession.
Planktonic foraminifera1 species are rare in the Coralline Crag when compared with the benthic foraminifera. The dominant species, N . arlantica, has a strong robust test and is also found in the higher energy Facies B (Mathers et al., 1984) at localities near Sudbourne, including Crag Farm (see Fig. 1); planktonic foraminifera of Facies C are too badly preserved, however, to be specifically identifiable (G. E . Hodgson, pers. comm.). All other species, with the exception of G . puncticulata, are very rare, with only a few specimens of most and with single specimens only of Orbulina universa d'Orbigny and Turborotalita cf. quinqueloba (Natland). The SEM photographs in Plates 1 and 2 show that the tests are well preserved.

Remarks.
The illustrated specimen has a broken-off bulla. The species has only been found at Ramsholt Cliff. A few small specimen<, however, were also found at Sutton Knoll and Gedgrave Cliff with n o trace of a bulla, and may belong to the possibly related species C. juvenilis Bolli (see comments in Jenkins et al., 1986); they are not illustrated.
Orford Borehole P 1983 Globorotalia puncticulata (Deshayes); King: 36, pl. 6, figs. 14, 15. Remarks. There is good evidence that G. puncticulata evolved from G . sphericomiozea Walters in the Early Pliocene (Jenkins, 1975;Scott, 1980;Kennett & Srinivasan, 1983) but Malmgren & Kennett (1981) regard its ancestor as G. conomiozea Kennett. This evolution took place in the mid-high latitudes of the oceans of the Southern Hemisphere, but the evolutionary event has not been recorded either in the North Atlantic, where G . puncticulata made a late entry in the Early Pliocene (Poore, 1978), or in the tropicalsubtropical belt such as the eastern equatorial Pacific (Jenkins & Orr, 1972).
At around the Early/Late Pliocene boundary, in the mid-latitudes of the S.W. Pacific, G. puncriculata evolved into G . irzflata (d'orbigny), and in the North Atlantic Poore (1978) reports that there is an overlap in their stratigraphic ranges in the lower part of the Late Pliocene of DSDP Site 410 (Fig. 2). Weaver (1986), on the other hand, states that there is no overlap in the ranges of the two species in the upper part of the Late Pliocene of the DSDP Sites drilled on Legs 80 (Site 548). 81 (Site 552A) and 94 (Sites 606-611) (Fig. 2). In the North Sea, King (1983) also recorded the extinction of G . puncticulata just before the first appearance of G . inflatu near the Early/Late Pliocene boundary. In the Coralline Crag, G. puncticulata is present but is unaccompanied by its descendent species, G . inflata.
One of the main morphological differences between G . puncticulata and G . infrata is the number of chambers in the final whorl. From counts made on S. W. Pacific populations, Malmgren & Kennett (1981) showed that this varies from a mean of 4.0 to 3.6 in G . puncticulata, in contrast to figures of 3.6 to 3.25 for G . inflata, the mean in each case decreasing with time from the earliest occurrence (first number) to the latest (second number). Of the 35 specimens of G. puncticulata obtained from the Sutton Knoll sample the following range of chamber counts in the final whorl was made: one specimen (4.5 chambers), 24 (4.0), eight (3.75) and two (3.5), giving a mean of 3.9 chambers in the final whorl.
In order to assess whether Malmgren & Kennett's (1981) figures hold true for populations in the North Atlantic, and are thus applicable to the Coralline Crag, an analysis was made of the chamber numbers in the final whorl of G . puncticulata at DSDP Site 609B (Fig.  2); 100 specimens were examined from each of four samples, one at the base of its stratigraphic range, one in the middle and two at the top. In the lowermost sample (Core Catcher 29; Early Pliocene), the mean number of chambers in the final whorl was 4.05, in the middle sample (Core Catcher 24) it was 3.99 and in the two top samples (Core Catcher 20 and 19; Late Pliocene) the figures were 3.97 and 3.93, respectively. Thus the results from both Site 609B and from the Coralline Crag would place these populations within the lowermost part of the stratigraphic range of G. puncticulata as given by Malmgren & Kennett (1981) for the S. W. Pacific. It therefore appears that in the North Atlantic this species was either undergoing a much slower rate of morphological change or it may not have been connected with the evolutionary lineage which led to G. influfa as reported from the Southern Hemisphere. Some evidence is provided for this latter hypothesis by Weaver (1986) who has reported a cryptogenic appearance of G . influfa at a number of DSDP Sites in the North Atlantic after the extinction of G. puncficulata.
Specimens of N. atfanfica were first recorded in the Coralline Crag by King (1983) but in the historical collections in the British Museum (Natural History) from Sutton, Gedgrave and localities around Sudbourne, it was previously identified as Gfobigerina bulloides d'Orbigny (see p. 2, herein). In the present study it was found at Ramsholt Cliff, Gedgrave Cliff and Sutton Knoll; it also occurs in the Coralline Crag samples from the Orford borehole in which Andrews & West (1977) identified pollen which they equated with the Brunssumian (Early Pliocene) pollen stage of the Dutch succession. In all the material so far examined by us the coiling direction is sinistral.
Ramsholt Cliff. then N. praehumerosa (Natori) is available, but here we take the view that it is not necessary.

Remarks.
Only one specimen was found in the Coralline Crag samples, from Gedgrave Cliff.

PALAEOECOLOGY
The duration of the Early Pliocene has been estimated at 1.8 Ma, from 5.2 to 3.4 Ma (Berggren et al., 1985) (Table 2). It was a time of high planktonic foraminifera1 diversity in the North Atlantic. At DSDP Sites 407 and 408 (lat. 63" N) (see Fig. 2) Poore (1978) recorded 25 and 22 species respectively and at the warmer Site 410 (lat. 45" N) there were no less than 45 species present. The low species diversity in the Early Pliocene sediments of the North Sea area, with five species listed by King (1983) and nine species found by us in the Coralline Crag (lat. 52" 30'-53" 00' N), is therefore relatively anomalous.
It is possible that the English Channel was not open in the Early Pliocene (Pomerol, 1982) and the low diversity fauna of the North Sea area was only connected to the North Atlantic via the northern passage (at about lat. 60" N). Comparison would therefore be expected with the fauna obtained from DSDP Sites 407 and 408 (Table 2; Fig. 2). Indeed, examination of the faunas from within the stratigraphic range of Globorotalia puncticulata (Deshayes) at these sites reveals a diversity of only 16 and 19 species, respectively, in the upper part of the Early Pliocene. On the other hand, the southern North Sea is sufficiently remote from open oceanic waters to account for the Explanation of Plate 2 All specimens are X125 Figs. 1-3. Neogloboquadrina pachyderma (Ehrenberg). Spiral, edge and umbilical views, P 51980. From Ramsholt  (Berggren): figs. 5-7, spiral, edge and umbilical view of a high-spired individual, P 51982; figs. 8-10, spiral, edge and umbilical views, P 51983. Both from Sutton Knoll. Cliff. 2080 lower diversity, even with a southern connection. The abscence of some of the deeper water species, such as Gfoborotalia scitula (Brady) in the Coralline Crag can be explained by its shallow water origin.

AGE AND CORRELATION
At DSDP Site 410 (Fig. 2), Gfoborotalia puncricufuta (Deshayes) first appeared in the lower part of the Early Pliocene (Poore, 1978) where its range overlaps that of G . conomiozea Kennett, its immediate ancestor in the S. W. Pacific, according to Malmgren & Kennett (1981). Further north at DSDP Sites 407 and 408 (Fig.  2), G. puncticulata made a later entry in the Early Pliocene and is unaccompanied by its immediate ancestor (Poore, 1978). Since it makes a cryptogenic appearance, it is therefore reasonable to infer that its entry into the North Sea may have been at a similar time to, or even later than at Sites 407 and 408 (assuming the connection to the North Atlantic was via the northern entrance to the North Sea). If this is accepted the placing of the entry of G. puncticufata into the North Sea at the Miocene/Pliocene boundary by King (1983) is probably too early. Weaver (1986), having examined the planktonic foraminifera of DSDP Sites 606 and 611 (Leg 94), considers that G. puncticulata appeared for the first time in the North Atlantic in the Early Pliocene at c. 4.2 Ma and that it became extinct just before the first appearance of its descendent species, C. inffata (d'Orbigny) ( Table 2). Poore (1978) showed a long overlap in the ranges of these two species at Site 410 and a possible short overlap at the more northern Site 408 and no overlap at Site 407.
Hooper & Funnell (1986) also recorded . . . "an apparent and very rapid transition between these two species, at around 2.3 Ma" at DSDP Site 552A (Fig. 2) which is at lat. 56" N in the North Atlantic, but later concluded that the first appearance of G. inffata was caused by its migration and not by its direct evolution from G. puncticulatu. Further south at DSDP Site 548 (Fig. 2) at lat. 48" N, Loubere & Jakiel (1985) concluded that there was an evolutionary transition between the two species, although their stratigraphic range chart for the site shows the overlap only occurring in one sample. Whereas Poore (1978) recorded the first appearance of C. inffata with the first ice-rafted material in the North Atlantic, dated by Shackleton et af. (1984) Table 2. Hodgson & Funnell (1986) have recently reported the occurrence of the (?) planktonic algal cyst, Bolboforma costura Murray, in a sample of Coralline Crag taken from an excavation near Ramsholt Cliff. This fossil is so far only recorded in the nannofossil Zone "15 (i.e. between c. 3.5 and 3.4 Ma) (Murray, 1984) and it seems probable that the Coralline Crag itself represents only the early part of the joint range of G. puncticufatu and N. atlantica.

CONCLUSIONS
The Coralline Crag deposits were laid down during the Pliocene within the joint ranges of C. puncticufata and N. atlantica (c. 4.2 to 2.3 Ma); other evidence (Hodgson & Funnell, 1986) suggests that the deposits are mostly Early Pliocene in age, probably no younger than c. 3.4 Ma.
As far as we can judge, all our samples containing planktonic foraminifera, including the Gedgrave Cliff samples taken from an excavation at the base of the section, are referable to Facies Arelatively low energy deposits underlying the higher energy, more generally exposed, sand-wave facies (Hodgson & Funnell, 1986) of the Coralline Crag. Tidal currents in the southern North Sea Basin may have been less strong and a southwestern connection with the North Atlantic may have been less well developed then than during later Coralline and Red Crag times. In any case the low planktonic foraminifera1 species diversity indicates a rather remote connection with the open ocean waters of the North Atlantic. The planktonic foraminifera suggest surface water temperatures i n the range 10-18" C .