9. PALEOGENE AND NEOGENE BENTHIC FORAMINIFERS FROM ROCKALL PLATEAU1

John W. Murray, Department of Geology, University of Exeter2

With a contribution by J. F. Weston

ABSTRACT

Sites 552 to 555 were drilled on a sediment-starved subsiding passive margin. The Cenozoic sedimentary successionshave yielded benthic foraminiferal assemblages of late Paleocene to Pleistocene age. Paleoecological analysis, usingcomparisons with data on modern assemblages, shows that the environments ranged from marginal marine in the latePaleocene to early Eocene to bathyal in the Oligocene and Neogene. Comparison with other DSDP results suggests thatfaunal communication between the Rockall area and the Norwegian Sea existed from the early Eocene and that cooldeep water entered the area in the late Eocene to early Oligocene. The early Miocene bathyal assemblages are clearlyrecognizable as of North Atlantic Deep Water type.

INTRODUCTION

Four sites (Holes 552, 552A; 553, 553A, 553B; 554,554A; and 555) were drilled on the sediment-starved pas-sive margin to the southwest of Rockall Plateau (Table 1,Fig. 1). Detailed studies of the benthic foraminifers werecarried out both on board Glomar Challenger and sub-sequently in Exeter, to interpret the subsidence and envi-ronmental history of the area. Previously, Sites 403 and404, drilled on Leg 48, had shown that this margin hadundergone subsidence from around sea-level in the latePaleocene to early Eocene to bathyal depths (>2000 m)by the Neogene. It was hoped that the use of the hy-draulic piston corer on Leg 81 would greatly improvethe quality of sampling and thus permit more detailedstudies to be carried out. This proved to be the case atHole 552A. All other sites were drilled using conven-tional methods but nevertheless a more complete recordwas obtained than during Leg 48. Sites 554, on the outerhigh, and 555, drilled to the north of Edoras Bank, haveadded much new information about previously undrüledareas.

An account of downhole occurrence of the benthicforaminiferal assemblages is given in each of the site chap-ters and is not repeated here. The primary data are listedin Tables 5 to 11. In this chapter comparisons are madebetween the sites, to draw out the similarities and differ-ences, to describe the faunas in a stratigraphic sequence,to interpret the conditions of deposition, and to discussthe implications of these results for the understandingof the geological history of the area.

METHODSThe Paleogene and the more compacted Neogene sediments were

dried in an oven and then soaked in a dilute solution of sodium hexa-metaphosphate. All samples were washed on a 63 µm (230 mesh) sieve,dried, and then split on a 125 µm (120 mesh) sieve. Only the fraction

Table 1. Position of sites, Leg 81.

Hole

552552A553553A553B554554A555

Latitude

56°02.56'N56°02.56'N56°O5.32'N56°05.32'N56°O5.32'N56°17.41'N56°17.41'N56°33.70'N

Longitude

23°13.88'W23°13.88'W23°20.61'W23°20.61'W23°20.61'W23°31.69'W23°31.69'W20°46.93'W

Depth ofwater (m)

23012301232923292329257625741666

Penetration(m)

314.0183.5

9.0682.533.576.0

209.0964.0

1 Roberts, D. G., Schnitker, D., et al., Init. Repts. DSDP, 81: Washington (U.S. Govt.Printing Office).

2 Addresses: (Murray) University of Exeter, North Park Road, Exeter EX4 4QE, UnitedKingdom; (Weston) Stratigraphic Services International (UK) Limited, Tannery House, Tan-nery Lane, Send, Woking GU23, 7EF, United Kingdom.

> 125 µm was examined. Counts of at least 100 benthic individualswere made except when the faunas were very sparse. The planktonic tobenthic ratio is based on a count of at least 100 individuals.

Altogether 17,628 benthic foraminifers have been mounted fromthe 162 fossiliferous samples studied. The stratigraphic assignment ofeach sample is based on the biostratigraphy determined from theplankton (nannoplankton, planktonic foraminifers, radiolarians, anddiatoms—see the site chapters, this volume).

The determination of the diversity index («) is based on the graphgiven by Murray (1973, p. 9) following the method of Williams (1964).

Computer analysis was carried out using Q mode varimax factoranalysis.

CRITERIA FOR INTERPRETING FOSSILASSEMBLAGES

Knowledge of the ecology and distribution of mod-ern benthic foraminifers provides the comparative basisfor interpreting the paleoecology of fossil assemblages.Although there is a large body of data on shelf and mar-ginal marine environments, the foraminiferal assemblagesof the slope and deep ocean are much less well known(Murray 1973; Boltovskoy and Wright, 1976; Douglasand Woodruff, 1981). This clearly limits the paleoeco-logical interpretation of fossil oceanic deposits.

There are numerous problems in the study of modernoceanic benthic foraminifers:

Abundance. In sediments from above the lysocline,planktonic foraminifers outnumber the benthic forms byat least 99 to 1. Thus, picking an assemblage of > 100individuals is very time consuming. Furthermore, theproportion of living forms is very low indeed. The low

503

J. W. MURRAY

Figure 1. Location of Sites 552 to 555, other DSDP sites used in discussion, and movement of Norwegian SeaOverflow Water into the North Atlantic. Long arrows based on Worthington (1970); short arrows based onAlvinerie et al. (1976, after Berthois and Auffret).

sedimentation rate results in dead tests swamping theliving. Further, because of the generally low fertility ofoceanic bottom water, the density of living individuals iscommonly low anyway.

Live versus dead. It follows that almost all studies ofdeeper water benthic foraminifers are based on dead as-semblages. These are bound to differ from the livingassemblages from which they are drawn because ofproduction differences between species and postmortemchanges such as mixing as a result of bioturbation ortransport (Murray, 1976a). Since the distribution of in-dividual living species on a scale of centimeters is likelyto be patchy, with the location of patches varying withtime, the diversity of the dead assemblages will be muchhigher than that of the living assemblages from whichthey are drawn. A 1-cm-thick sample from an area hav-ing a 2 cm/1000 yr. sedimentation rate will represent 500years and therefore many generations of foraminifers.

Correlation with ecological parameters. In ecologicalstudies it is normal practice to try to explain distributionpatterns with respect to limiting parameters. There areparticular problems in trying to do this for deeper waterforaminifers.

Physical and chemical oceanographers have recognizedthat there are different water masses in the oceans, eachhaving its own source, its own direction of movement,and its own physical and chemical attributes. However,the differences are often small: salinity may vary by afew decimal points in parts per thousand; temperaturedifferences may be fractions of a degree. An essentiallyunanswered question is whether benthic foraminifers arereally able to discriminate such small differences. Thesewater mass measurements are not usually made at thesediment/water interface but some distance above, soare they truly representative of the environment in whichthe foraminifers live? It may not be so for such parame-

ters as dissolved oxygen and CaCO3 availability. Themost easily measured and most obviously changing pa-rameter is water depth, but most other environmentalparameters are interrelated with depth. However, in theNorwegian-Greenland Sea, Belanger and Streeter (1980)found nonvarying physical and chemical water parame-ters at depths greater than 1500 m although the benthicforaminifers showed marked assemblage changes. Thisled them to observe that, "Under such situations of anoverall stable environment, it appears that the deep-seabenthos exhibit a sensitivity to parameters we are notused to looking at and that affect species relative andabsolute abundance."

The concept of limiting parameters is that each spe-cies has its own particular tolerance range for each fac-tor. When any one factor exceeds the tolerance limit forthat species, local extinction will occur. In some casestwo or more interrelated factors control local extinctions.It follows that no single limiting factor will be responsi-ble for causing local extinction.

Faunal changes through time. It is now known thatthere have been major changes in the distribution of deep-sea benthic foraminifers since the last glacial maximum.The changeover in the oceans proceeded from the topdownwards at depths greater than 2500 m in the westernNorth Atlantic and was a diachronous event (Schnitker,1979b; Streeter and Lavery, 1982). At individual coresites certain species were more abundant during the gla-cial event than they are now, e.g., Uvigerinaperegrina,U. canariensis; others are now more abundant, e.g.,Hoeglundina elegans, Pyrgo murrhina, Oridorsalis um-bonatus, Osangularia rugosa, and Epistominella exig-ua. However, the emphasis is on changes in abundance;no new species have been introduced and none has be-come totally extinct. Perhaps the most significant aspectis that these changes are most readily correlated with

504

BENTHIC FORAMINIFERS

changes in water masses and their circulation ratherthan just with depth.

Published Work. The most recent review of moderndeep sea benthic foraminiferal distributions is that ofDouglas and Woodruff (1981). Although many deep wa-ter forms are cosmopolitan, there are differences in dis-tributions not only between but also within oceans. Forthe interpretation of Rockall material, comparative datafrom the North Atlantic have been used.

Phleger et al. (1953) have provided much informationon distributions in the whole North Atlantic. This, to-gether with data from additional samples, has been sub-jected to computer analysis by Streeter (1973), whileSchnitker (1974) presented new data from the westernpart only. Both authors found a good correlation be-tween assemblages and water masses. On a regional scalethe water masses are transgressive. Schnitker (1980) sum-marized the relationship for the North Atlantic (Table 2).

The named species are dominant and are accompa-nied by many associate species. Most species are presentin more than one water mass.

Streeter and Lavery (1982), in a study of the conti-nental slope and rise between Cape Hatteras and Tail ofthe Banks, found that four associations are closely re-lated to depth. Taking data from Miller and Lohmann(1982) for Cape Cod, and from Schafer and Cole (1982)for Newfoundland, it can be seen that this depth-relateddistribution extends over a great distance (Table 3). TheUvigerina assemblage is absent off Newfoundland butthe Hoeglundina assemblage starts at the same depth.However, with the exception of the study by Schafer andCole (1982), all these papers lack details of the assem-blages as a whole, and they are therefore difficult to ap-ply to the fossil record. A further complication is thatcertain species common in the Rockall area appear to beabsent from the western Atlantic, e.g., Bulimina alaza-nensis, Cibicidoides kullenbergi, and Triloculina frigida(Culver and Buzas, 1980, listing consulted).

It should also be noted that the foraminiferal assem-blages of the Labrador Sea are dominated by large num-bers of agglutinated individuals (Phleger, 1952; Schaferand Cole, 1982) and in this respect are very differentfrom those of the Rockall area.

Deep Water Benthic Foraminiferal Assemblages fromthe Recent NE Atlantic

Contribution by J. F. Weston

Work in progress on Recent benthic foraminiferal fau-nas from the NE Atlantic has revealed the existence of avery diverse foraminiferal fauna. The following species

Table 2. Water masses and foraminiferal faunas of the North Atlantic

(based on Schnitker, 1980).

Table 3. Distribution of benthic foraminiferal assemblages on the con-

tinental margin of North America. (Data from Streeter and

Lavery, 1982; Miller and Lohmann, 1982; and Schafer and Cole,

1982.)

Water mass Western Eastern

Antarctic Bottom Water (AABW)Arctic Bottom Water (ABW)North Atlantic Deep Water (NADW)

Lower

Upper

O. umboniferaE. βxigua

E. exigua

U. peregrina

O. umbonifera—

P. wuellerstorfi

U. peregrina

Depth (m)

1000

2700

3900

Cape HatterasU t . 35°N

Bulimina assemblage:B. marginata, B. aculeataGlobulimina affinisBrizalina spp.Trifarina williamsoniCassidulina crassa

Uvigerina assemblage:Uvigerina peregrina

Hoeglundina assemblage:Hoeglundina elegansCibicidoides wuellerstorfiOridorsalis umbonatusGlobulimina affinisMelonis pompiloides

Osangularia assemblage:Osangularia umboniferaCibicidoides wuellerstorfi

Tail ofthe Banks

45°N

1800

2600

3200

Newfoundland50°N

Planulina wuellerstorfiPyrgo murrhina

Hoeglundina elegansEpistominella exiguaPlanulina wuellerstorfiLimit of survey

show some depth restriction, especially within the areafrom 50-60°N latitude:

Bolivina aff. B. thalmanni: 1500-3000 m, but is mostimportant in samples from 4000 m: Not commonin Recent sediments from the Rockall-Reykjanes Ridgearea, and more abundant in samples from south of 50°N.It is absent from core-top samples from Sites 552-555,but is present in the Neogene of these sites.

Cibicidoides kullenbergi: >1500 m: Rare in Recentsediments north of 50°N, but abundant in samplessouth of this at depths between 3500 m and 4200 m.

Epistominella exigua: 1500 m->5000 m: Probably themost common species in the deep waters of the NE At-lantic. It often makes up 20% of an assemblage in sam-ples from depths of 2500-3500 m.

Melonis pompilioides: Consistent upper depth limitof 2000-2200 m, with rare shallower occurrences.

Oridorsalis umbonatus: 1500-> 5000 m: Maximumabundances of this species occur at >2500 m.

Planulina wuellerstorfi: Consistent upper depth limitof 1000-1200 m, but most abundant at depths >2000 m.

Sigmoilopsis schlumbergeri: 500-3500 m, but whenabundant, the depth is always

J. W. MURRAY

storfi = P. wuellerstorß). However, T. frigida does occurrarely in samples from the Norwegian-Greenland Sea(Belanger, pers. comm. to J. W. Murray).

Two other individual species show relationships withexternal factors:

Brizalina subaenariensis: Occurs in submarine can-yon areas and regions of organic-rich deltaic sediments(Lutze, 1980). It has been interpreted by Sen Gupta etal. (1981) as indicative of localized upwelling, and in theNE Atlantic it appears to have a distribution related toareas of rich food supply.

Melonis barleeanus: Occurs throughout the NE At-lantic, but is particularly abundant in areas where thereare active bottom currents.

Qualitative data from the Recent samples have beensubjected to Q-mode varimax factor analysis. Four outof ten assemblages derived from this analysis are impor-tant in open ocean areas of the NE Atlantic north of40°N. The major species components of the factor as-semblages are shown in Table 4, and the Recent distribu-tion of these assemblages is shown in Figure 2. (The fac-tor distribution is drawn up from those factors which re-cord loadings of 0.5 to 1.0 or -0.5 to -1.0 in eachsample.)

The factor which is most important in samples from1500-3000 m in the area north of 50°N in the Recent

Table 4. Modern assemblages represented by varimax factors(factor scores × 100).

Varimax factor 1

50.4 Globocassidulina subglobosa50.3 Planulina wuellerstorfi37.1 Oridorsalis umbonatus34.2 Cibicidoides kullenbergi22.3 Epistominella exigua21.1 Complex agglutinants

Varimax factor 4

-86.0 Epistominella exiguaHoeglundina elegansBolivina spp./Brizalina spp./Fursenkoina spp.

Varimax factor 8

- 60.6 Globocassidulina subglobosa- 50.9 Osangularia rugosa- 29.3 Bolivina spp. /Brizalina spp. /Fursenkoina spp.-20.7 Uvigerina spp.

Varimax factor 10

Pullenia spp.Astrononion spp./TV. irideaBolivina spp./Brizalina spp./Fursenkoina spp.Bulimina spp./Francesita spp.Uvigerina peregrina

-25.7-22.2

-47.3-44.2-31.0-21.0-19.9

50'

40°

Key• Sample points

Φ Multiple sample points

X Factor 1l j Factor 4V.//J Continental margin and shallow-water factors8,10 Factors 8, 10

Figure 2. Distribution of Q-mode varimax factor modern assemblages in the northeast Atlantic.

506

BENTHIC FORAMINIFERS

North Atlantic is factor 4. This represents an assemblagedominated by E. exigua. Hydrographic profiles from theErika Dan cruise (Worthington and Wright, 1970) indi-cate that these samples (Fig. 2) are taken in areas over-lain by Labrador Sea Water and Norwegian Sea Over-flow Water.

The true North Atlantic Deep Water (NADW) is com-posed of Norwegian Sea Overflow Water mixed with con-siderable quantities of less saline Denmark Straits Over-flow Water (Worthington, 1976). The Norwegian SeaOverflow Water present in this area of the NortheasternAtlantic is referred to by some authors as "NortheastAtlantic Deep Water" (NEADW) (e.g., Lonsdale andHollister, 1979). NEADW forms an upper layer of theNADW (s.l.) which may be concentrated toward theeastern margin of the European Basin. The factor 4 as-semblage shows a distribution in the Recent NE Atlanticthat is very similar to the distributions of Labrador SeaWater and NEADW.

South of 50°N, samples from > 3500 m are charac-terized by factor 1 assemblages. This is a more diverseassemblage than that represented by factor 4 (see Table 4)and is similar to assemblages related to the distributionof NADW in other areas (e.g., Lohmann, 1978). Hydro-graphic data for areas where factor 1 shows the highestloadings indicate that, at the depths of these samples,the bottom water mass is probably NADW (s.s.) (U.S.National Oceanographic Data Center Files).

Factors 8 and 10 have relatively restricted occurrencesas the important factors in a sample. Factor 8 is impor-tant only at 2000-2800 m on the slopes of seamounts.Factor 10 is important only in one sample from 2353 mon the eastern slope of the Reykjanes Ridge. It may berelated to the relatively high salinity southward flow ofNorwegian Sea Overflow Water which occurs at this depthin this area (Worthington, 1976).

Therefore, these varimax factor assemblages appearto have distributions related to the distribution of bot-tom water masses in the Recent NE Atlantic Ocean. Thesefactor assemblages and their distributions are differentfrom those outlined in Streeter (1973) and Schnitker(1974, 1980) and indicate differences between foraminif-eral faunas in the northeastern and northwestern Atlan-tic (possibly enhanced by the examination of different sizefractions).

This is confirmed by the comparison of the distribu-tions of individual species in the northwest and north-east Atlantic, Labrador Sea, and Norwegian Sea, alreadydiscussed.

PaleogeneMuch is known of the distribution of genera and spe-

cies and the character of assemblages of modern benthicforaminifers of shelf seas and marginal marine environ-ments (Murray, 1973; Boltovskoy and Wright, 1976). Thepaleoecological interpretation of such assemblages fromthe Cenozoic is readily made on such criteria as diversi-ty, triangular plot of the suborders, genera, and plank-tonic to benthic ratio (Murray and Wright, 1974; Mur-ray, 1966).

The recognition of deeper water assemblages is muchmore difficult for several reasons: The study of moderndeep water forms is still in its infancy. There is evidencefrom the Pacific Ocean that some Paleogene assemblageshave changed their depth preferences with time, somemigrating into shallower and others into deeper water(Douglas and Woodruff, 1981); in the Rockall Plateauarea many of the assemblages are poorly preserved as aresult of dissolution.

A useful technique to establish the depth of accumu-lation of a fossil assemblage overlying ocean crust is to"backtrack" the site, that is, correct the depth for a par-ticular age by subtracting the amount of subsidence andadjusting for sediment thickness (Berger and Winterer,1974; Douglas and Woodruff, 1981). However, when itis desired to use the foraminifers to determine the depthof deposition, and thereby the subsidence curve, such atechnique cannot be used.

Using published data, the assignment of assemblagesto different depths has been summarized for the NorthAtlantic in Figure 3. Although based on comparisonswith modern assemblages as well as with other inverte-brate groups, such depth assignments are inevitably sub-jective. Shelf assemblages often contain Anomalinoideshowelli as a common component with Cibicidoides alle-ni, C. succedens, Alabamina obtusa, Pulsiphonina pri-ma, and Lenticulina spp. The diversity is moderate andthe planktonic to benthic ratio varies from 0 near theshore to 40-50% planktonic forms on the outer shelf.

The slope species include Nonion havanensis, Nuttal-lides truempyi, and Oridorsalis ecuadorensis, althoughoften not in great abundance. Such species also extendinto outer shelf environments. Much of the slope fauna ismade up of Nodosaria/Stilostomella spp., Gyroidinoidesspp., Cibicidoides spp., and Pullenia spp. Diversity ismoderate to high (oc > 15), and the proportion of plank-tonic tests exceeds 50% unless affected by dissolution.

QUATERNARY AND NEOGENE ASSEMBLAGESThe numerical results are listed in Tables 5 to 7, and

the more important general attributes are shown onFigures 4 to 9. The assemblages are dominated by Rota-liina, with Textulariina and Miliolina playing very sub-ordinate roles. The diversitv is moderate to high with ocvalues in the range 10 to 30. Because of this, individualspecies do not normally form a very high proportion ofthe total.

Pleistocene and "Glacial" Late PlioceneThe best record is that of the HPC of Hole 552A

(Cores 1,CC to 9,CC). This "glacial" part of the succes-sion is marked by cyclic changes from white foram nan-no oozes to terrigenous oozes. Most of the core-catchersamples came from the white oozes representing inter-glacial conditions. The diversity shows marked fluctua-tions from

J. W. MURRAY

Meters Eocene Oligocene

Protelphidium

5 0 -

100-

150-

200-

Anomalinoides howelliCibicidoides alien iCibicidoides succedensOsangularia e×pansa

250-

500-

Albamina, LenticulinaAnomalinoides howelliCibicidoides alien/Cibicidoides succedensPulsiphonina prima

Alabamina, LenticulinaAnomalinoides howelliGlobocassidulina subglobosaGyroidina angustiumbi/icataOsangularia e×pansa

Alabamina, GyroidinaAnomalinoides howelliOridorsalis ecuadorensisLenticulina

Osangularia pteromphalia, Gaudryina hiltermanniGavelinella grosserugosa, Gyroidina angustiumbi/icataOridorsalis ecuadorensis, Melonis affinis

Gyroidina, GyroidinoidesLenticulinaNonion cf. N. olssoniOridorsalis ecuadorensisSpiroplectammina spectabilis

Cibicidoides, NodosariaGyroidinoides complanataNonion cf. N. olssoniOridorsalis ecuadorensis

Nuttallides truempyi, OsangulariaOridorsalis ecuadorensis, CibicidoidesAnomalindoides grosserugosa

1000-Stilostomella, Gaudryina hiltermanniSpiroplectammina spectabilisNuttallides truempyi, OsangulariaOridorsalis ecuadorensis Stilostomella

PleurostomellaNuttallides truempyi

Bulimina trigonalisNodosaria, StilostomellaOridorsalis ecuadorensisNuttallides truempyi 2

Oridorsalis ecuadorensisCibicidoides, StilostomellaGyroidinoides girardanusSiphonina tenuicarinataHeterolepa me×icanaAnomalinoides grosserugosa

Figure 3. Depth distribution inferred for Paleogene assemblages in the North Atlantic. (1. Berggren, 1974; 2. Murray, 1979; 3. Berggren and Aubert,1976b; 4. Berggren and Aubert, 1976a.)

Pyrgo murrhinα. Three species are restricted to this in-terval: Cαssidulinα teretis, which appears in abundancein 2,CC and 9,CC; Triloculinα frigidα, which is presentfrom 1,CC to 5,CC (i.e., only in the Pleistocene); andCibicides sp. Melonis pompilioides, which is present inevery sample but is rare in the preglacial Neogene. Glo-bocαssidulinα subglobosa is present only rarely in Core6,CC but is common in the preglacial Neogene. Cibici-doides kullenbergi is intermittently present in the glacialsuccession and invariably present below it.

The record at the other sites is far less complete butthe pattern of occurrence is essentially the same, exceptthat at Site 555 Melonis pompilioides is rare throughoutthe whole Neogene because of the shallow depth of thesite.

Preglacial Pliocene and Late Miocene

In Hole 552A the fauna comprises the commonlyoccurring forms Epistominella exigua, Globocassiduli-na subglobosa, Gyroidinoides spp., Nodosaria/Stilo-stomella, Oridorsalis umbonatus, and Planulina wuel-lerstorfi. Two species are more abundant than they arein the succession above: Cibicidoides kullenbergi andUvigerina compressa. Ehrenbergina trigona is restricted

to this interval while E. serrata and Laticarinina pauper-ata range down into the middle Miocene. Bulimina ala-zanensis and Brizalina subaenariensis also extend throughthis interval.

The commonly occurring forms are also present atthe other three sites. C. kullenbergi and U. compressashow the same trend in abundance as at Hole 552A.

Middle Miocene

This part of the succession is best represented in Hole553A (6,CC to 7,CC). The fauna is essentially the sameas that of the late Miocene, but Ehrenbergina trigonahas not been recorded in this interval. Siphonina tenui-carinata is present in Holes 554A and 555.

Early Miocene

This is incompletely represented in all four sites. Thecommonly occurring species are Cibicides sp., Cibici-doides kullenbergi, Globocassidulina subglobosa (espe-cially in Sample 555-24-7, 55 cm), Gyroidinoides spp.Melonis barleeanus, Nodosaria/Stilostomella, and Ori-dorsalis umbonatus. One notable absentee is Planulinawuellerstorfi. Holes 552A, 553A, and 555 have Siphoni-na tenuicarinata, while Alabamina sp. is abundant andOridorsalis ecuadorensis is present in Hole 553A.

508

BENTHIC FORAMINIFERS

Table 5. Benthic foraminiferal census data, Hole 552A, Neogene.

Species

Astrononion guadelupaeBolivina cf. B. thalmanniBrizalina suboenariensisBrizalina sp.

Bulimina alazanensisBulimina striataCassidulina obtusaCassidulina tsretisCibicides sp.Cibicidoides bradyiCibicidoides kullenbergiCibicidoides robertsonianusCibicidoides sp.Dentalina sp.Eggerella bradyiEhrenbergina serralaEhrenbergina trigonaEponides tumidulusEpistominella exiguaFissurina, Lagena, OolinaFrancesita advenaFurnkninfi srhrpihprsiana

Globocassidulina subglobosaGlobulina, GuttulinaGyroidinoides spp.Hoeglundina elegansKarreriella bradyiLaticarinina pauperataLenticulina sp.Melonis barleeanusMelonis pompilioidesMelonis sp.Nodosaria, StilostomellaNonionella sp.Oridorsalis umbonatusOsangularia rugosaPlanulina wuellerstorfiPleurostomella spp.Pullenia bulloidesPullenia osloensisPullenia quinquelobaPullenia salisburyiPyrgo bulloidesPyrgo murrhinaPyrgo sp.Quinqueloculina sp.Rectuvigerina royoiSigmoilopsis schlumbergeriSiphotextularia catenataSphaeroidina bulloidesSiphonina tenuicarinataSpiroloculina pusillaTriloculina frigida

Uvigerina compressaUvigerina peregrinaVulvulina pennatulaUnidentified

Total

Number of speciesRatio planktonic to

benthic (x:l)

33

'I

_

8

14

—

1

\

13

—

4

_

22

7—

_2

6

—

1—

12

8———

62

13—

37

—11

3

1

11

—

1η

2

13

116

46

99

2,C

C—

1

14

511

3—

_

3

—

15

5

3

2—

1—

—

5—

1

1

5

2

9

2

2_

—

—

3

1

1——

1

11

5

133

34

49

Pleistocene

3,C

C

—

2

14

1

_

—

2

56

3

—

41

—_

—

211

—

552

14

10—

1—

2

—

1

1

_

12

2

6

111

3199

U

u

54

13g1

—

—

227

9

—

15

—_

1

1—

1

25

_

12

34

—

—

1

1

1——

_

5

—

104

30

99

U

q

—

8

_

3

_

2

_

52

—_

6

1

6

—_

—

31

9_

1—

11

1_

16

1—

83

—

2

1

η

42

7

102

3899

6,C

C

10—

_

_3

3

5j

_

—

1

1

4

51

8

1—

1

2

7

15—

12—

324

2

1_

1

4

1

—

——

1

1

1

3

106

39

99

7,C

C

34

_

_

45

2

4

8

12

z6

1_

1

67

_

8_

282127

5A

1

3

_

3

109

4099

8,C

C

_

7

7

_

3

π8I

9

7

1_

_

3

7

1

212

2_

21

5

2

31

1

4

102

3899

33'6

1

1_

35

2_

_

5

1014

—_

144

55

7_

3

_

7

1

_

z

_3

—

102

2749

latePliocene

U

10

2

4

_

22

4

121

61

_11

z35

—

1

2_

811

_

6

7

31

_

_

1102

2_

1

8

2

109

42

99

J. W. MURRAY

Table 6. Benthic foraminiferal census data, Holes 553, 553A, 554, 554A, Neogene.

Species

Alabamina sp.Astrononion guadelupaeBolivina cf. B. thalmanniBrizalina subaenariensisBrizalina sp.

Bulimina alazanensisBulimina striataCassidulina obtusa

Cibicides sp., Cibicidina sp.Cibicidoides bradyiCibicidoides kullenbergiCibicidoides robertsonianusCibicidoides sp.Dentalina sp.Eggerella bradyiEhrenbergina serrataEhrenbergina trigonaEpistominella exiguaEponides tumidulusFissurina, Lagena, OolinaFrancesita advenaFurKpnkninπ vrhrpihprstπnπ

Globocassidulina subglobosaGlobulina, GuttulinaGyroidinoides spp.Hoeglundina elegansKarreriella bradyiLaticarinina pauperataLenticulina sp.Melonis barleeanusMelonis pompilioidesMelonis sp.Nodosaria StilostomellaOridorsalis ecuadorensisOridorsalis umbonatusOsangularia rugosaPlanulina wuellerstorfiPleurostomella spp.Pullenia bulloidesPullenia osloensisPullenia quinquelobaPullenia salisburyiPyrgo murrhinaPyrgo sp.Quinqueloculina sp.Rectuvigerina royoiSigmoilopsis schlumbergeriSiphonina tenuicarinataSiphotextularia catenataSphaeroidina bulloidesSpiroloculina pusillaSpiroplectammina spectabilisTriloculina frigida

Uvigerina compressaUvigerina peregrina

Vulvulina pennatulaUnidentified

Total

Number of speciesRatio planktonic to

benthic (x:l)

Pleist.

Hole553

U

q

_—

2_

_

1

—

2

11

1

—

42

11

13

2

4

5

_

—

1

4—

—

2

3

2—

2

20—

5——

———

—

2——

5

214

5

109

38

99

latePliocene

U

q-_——_

2

2

_

—

1

2—

3

2

2

7—

71

10

2—

—

2

1314

11

2

92

47

—

7

2—

———

2—

—

_

8

1

125

37

99

Uu

—2

2_

12

1

_

—

4

_

3

2

2

2

2

6—

52

8

3

1—_

24

10

24

11

6

71

—

5——

———

5—

1

_

69

4

150

41

99

e.Plio.

U

q

_l411

1

_

—

8

1

3

2

2—

7

1

52

15

1—

—_

4

16

7

2

6

52

5

6—

———

———

4—

2

_

6—

—

118

36

99

1.Mio

Uq

ll

2

31

1

1

_

1

4

4

9

7

1

1_

1_

154

1

37

21

128

——

——

_

_

_i

53

_

—

100

3499

m.Mio

Hole553A

O

q

52

_

1

2

4

_

_

2

7

1617

2

_

57

11

414

13

207

1—

_

—

_

5

12

132

3399

q

—34

2

_

4

_

1

1

1

3

19

12

_

9

11

5

11

2247

_

1——

1_

6

_

1

105

2999

yq

1

2

2

7

1

1

1

—

5

5

5

81

17

2

5_

920

13

17

2

12

5_

3

1

2

2

2

3

_

_

2

156

41

32

y000

2111

2

1

3

_

_

—

1

202

10

_

3

_

14

14

6

3

10

6—

———

—

3

1

_

10

132

31

24

e.

Miocene

σ\

_T

o

—

_

1

1

1

134

3—

11

—

—

2—

12

j

33

—

820

32

—

55

4—

—

—

——

—

1

_

—

8

102

2899

Note: Dash = absent in assemblage count.

ern results to the fossil record. The interpretations canthen be judged as to whether or not they are reasonable.

The assemblages in Holes 552A, 553A, and 554, 554Aare all basically similar in that they are of the Planulinawuellerstorfi type. The commonly occurring species areCibicidoides kullenbergi, Epistominella exigua, Gyroi-dinoides spp., Melonis spp., Nodosaria/Stilostomella,Oridorsalis umbonatus, and Pullenia spp. The diversityis moderate to high (oc 10-30) and so too is the plank-tonic to benthic ratio (except in a few samples that havebeen affected by dissolution).

The P. wuellerstorfi fauna indicates depths > 1500 m.With the exception of the "glacial" succession in Hole552A, Melonis pompilioides is present in low abundance

and this may indicate depths >2000 m, while the gen-eral rarity of Sigmoilopsis schlumbergeri suggests depths>2200 m (its lower limit of distribution between 50°and 60°N). Likewise, E. exigua normally forms >20%of the assemblage at depths > 2500 m, but only rarelydoes it approach such high abundance at any of thesites. Thus, from these observations, it appears likelythat the depth of deposition of the middle Miocene toRecent in Holes 552A, 553A and 554, 554A was 2200 to2500 m (cf. present depths of 2301 m, 2329 m, and2576 m, respectively) (see Figs. 4-6, 8).

The assemblages at Site 555 are also of P. wueller-storfi type with common Brizalina subaenariensis, Buli-mina spp., Cibicidoides kullenbergi, Epistominella ex-

510

BENTHIC FORAMINIFERS

Table 7. Benthic foraminiferal census data, Hole 555, Neogene.

Species

Astrononion guadelupaeBolivina cf. B. thalmanniBrizalina subaenariensisBrizalina sp.

Bulimina alazanensisBulimina striataCassidulina obtusa

Cibicides sp., Cibicidina sp.Cibicidoides bradyiCibicidoides kullenbergi

Cibicidoides sp.Dentalina sp.Eggerella bradyiEhrenbergina serrataEhrenbergina IrigonaEpislominella exiguaEponides tumidulusFissurina, Lagena, Oolina

Globocassidulina subglobosaGlobulina GuttulinaGyroidinoides spp.Karreriella bradyiLaticarinina pauperataLenticulina sp.Melonis barleeanusMelonis pompilioidesMelonis sp.Nodosaria, StiiostomellaOridorsalis umbonatusOsangularia cullerOsangularia rugosaPlanulina wuellerstorfiPleurostomella spp.Pullenia bulloidesPullenia osloensisPullenia quinquelobaPullenia salisburyiPyrgo murrhinaPyrgo sp.Quinqueloculina sp.Rectuvigerina royoiSigmoilopsis schlumbergeriSiphonina tenuicarinataSiphotextularia catenataSphaeroidina bulloidesSpiroplectammina spectabilisTriloculina frigidaUvigerina auberianaUvigerina compressaUvigerina peregrinaVulvulina pennatulaUnidentified

Total

Number of speciesRatio planktonic to

benthic (x:l)

Pleist.

33

'I

1

3

14

1

1

1—

3

9—

—

6

2

31

——

1—

3

14—

—

2—

1

133

—

1

_

—

2—

—

4

9

104

3599

g31

_

11

1

5η

2

2

14

—

—

21

10

51

_

1

1—

14_

12151

1

j

_

1

_

1

2—

—

9

1

113

3899

e.PI.

3,C

C

221

851

135

—

—

—

-

7

512

—

1563

_

14

2212

—

4

—

_

—

4

—

103

3399

4,C

C

1

3_

541

18

14

131

—

2—

o

5

6

2

2

294

_

—

2_

3

16

_

1

5—

16A

6

127

4299

5,C

C

4

31

2152

1

3—

231QO

7

12

_

—

1512

1_

—

1

5

411

32

_

3

2—

ΛA

1

118

3599

6,C

C

101

_

3

101

_

13

13

_

32

£

7

7

11

194129

1

1_

1

1

1

1—

_

1

13

115

3699

7,C

C

_

_

3

66

_

_

66

2

5_

53

-

7

14

1—

1

14

8

1

4

_

22

_

4

1

7

112

3449

g4

13

172

_

_

23

_

12

_

69

-

9

7

11

644

—

37

52

_

1

—

—

1

_

2

109

3499

late Miocene

33'6

21321

_

2_

_

47

1

_

3

-

16

6

_—

5

311

5

46291

1

1

__

3

6

121

3199

ü

10,

_

322

19

_

_

65

_

2

—-

20

5

1

4

10145

_

1528

1

—2

1

2

4

120

3499

J. W. MURRAY

Pleist.

latePlio.

20/21

19

_1J_

16

15

earlyPlio.

50-

13/14

100

• 12

lateMioc.

m.Mioc.

11

150-

9/10

e. Mioc.

midEoc.

^7-9

171η

13

14/16 1 7 5

14/15

180-

- Ii co

"– 1,CC

• 3.CC

• 5,CC

7.CC

9,CC

-11,CC

-14.CC

-16,CC

-18,CC

-20.CC

-22.CC

-24.CC

-26.CC

-28,CC

-30,CC

-32.CC

-34.CC

lla

IV

2,CC

4,CC

6,CC

8.CC

10.CC

12,CC

Inferred waterdepth (km)

1 2 1 3

Plankton

100 0

Benthicspecies (α)

10 20 30 1

Varimax factorloadings

0 0.5 0 -0.5 -1.0

•I

H F K 35.CCIMP te ' 36-2, 100

•~~»-i ~-171 •v™—i

oiig. _ M _ _ ]πit:g?;i2 _£_„

BENTHIC FORAMINIFERS

Plankton Benthic

species (α)

100 0 10 20

Diss'n

Varimax factor

loadings

1-0 0.5 0 -0.5 -1.0

\ I

\

I

!

\ "

•

IV, V

IV, V•

Figure 5. Hole 552. Biostratigraphy based on nannozones from Backman (this volume), lithological unitsfrom site chapter (this volume), and inferred water depth at the time of deposition, determined fromthe benthic foraminifers. ( = present site depth.)

bergi, and Astrononion guadelupae. In Hole 554A, Spi-roplectammina spectabilis and Nonion havanensis are ad-ditional common species.

Most of these assemblages are from condensed se-quences that have been affected by dissolution. Preser-vation is commonly poor and the planktonic to benthicratio is often low because of the preferential loss of plank-tonic forms.

Late Eocene

Absent from all holes except 554A, where the faunais dominated by Nodosaria/Stilostomella spp., are Glo-bocassidulina subglobosa and Oridorsalis ecuadorensis.The diversity is moderate (oc9-13) and the planktonic tobenthic ratio low as a result of dissolution.

Middle Eocene

In Holes 552, 552A, and 553A the dominant speciesare Alabamina wilcoxensis, Bulimina tuxpamensis, No-dosaria/Stilostomella spp., Cibicidoides spp., Gyroidi-noides spp., Nuttallides spp., and Osangularia sp. Dis-solution has affected these assemblages to a greater orlesser extent, some levels being almost barren (e.g., Hole552: 9,CC and Hole 552A: 37-2, 137 cm; 37,CC; 38-1,62 cm; and 38,CC). With the exception of 553A-9-6,

51 cm to 553A-10-3, 66 cm, which have a planktonic tobenthic ratio of 99:1, the remainder have low values,probably mainly because of the effects of dissolution.

Early Eocene-Late Paleocene

Assemblages of this age were recovered from all foursites. Eleven different assemblages are recognized accord-ing to the environment of deposition:

1. Osangularia spp. dominant with Nodosaria/Stilo-stomella spp. subsidiary, but affected by dissolution(555-27-2, 105 cm and 28-1, 37 cm).

2. Anomalinoides howelli dominant with various ac-companying subsidiary species: Pulsiphonina prima and/or A. nobilis (552-12,CC to 21-2, 117 cm; 555-31-1,20 cm to 32,CC); Gavelinella semicribrata, Cibicidoidesspp., Lenticulina spp., and Elphidium hiltermanni;(553A-11-2, 80 cm to 16,CC); Anomalinoides sp., E.hiltermanni, Nonion laeve, and Pararotalia curryi (553A-18,CC); Protelphidium sp. (553A-19,CC).

3. Gyroidinoides spp. dominant, with subsidiaryAnomalinoides acutus, Lenticulina spp., Nodosaria spp.,Eponides spp., Cibicidoides alleni, Alabamina obtusa,and Praeglobobulimina ovata (555-44,CC; 45,CC).

4. Nodosaria/Stilotomella spp., Lenticulina spp.dominant, with subsidiary Praeglobobulimina ovata,

513

J. W. MURRAY

m

Pleist. NN20-21

l a t e NN16-18Plio.

100-

PNo. N N 1 6 - 1 8

early Plioc. NN12-15

late Mioc. NN9-11

200-

early Mioc. NN1-j Nf25_

I mid Eoc. NP16

/ - „ NP13NP149 ' N P 1 1 N P 1 2

early Eoc.300-

NP10

7 400-

late Paleo.

?NP9500-

Q

| . |

3 3

I

l la

lib-\-

111

IVa

rvb

IVc

IVd

IVe

IV f

Inferred waterE depth (km)

co 0 1 2 |

-1,CC

A-1.CC

• 2,CC

-3,CC

• 4,CC

- 5,CC

-6,CC

- 7,CC

-8.CC

-9-5,5V9,CC-10.CCb-11-6.52

vπ

f////•\\>/<V 7

/

/

/

/

T/

\\\

<{

j? I

s

•

<

II *

•

10

0

8

>

\

< I V

s

IV• I I I

Figure 6. Holes 553, 553A. Biostratigraphy based on nannozones from Backman (this volume), lithologi-cal units from site chapter (this volume), and inferred water depth at the time of deposition, deter-mined from the benthic foraminifers. ( = present site depth.)

Anomalinoides acutus, Alabamina obtusa, Cibicidoidesalleni, and Pullenia sp. (553A-2O,CC; 21,CC; 23,CC to37-2, 28 cm).

5. Lenticulina spp. dominant with subsidiary Alabam-ina obtusa and Anomalinoides nobilis (555-47,CC; 48,CC).

6. Praeglobobulimina ovata dominant with subsidi-ary Anomalinoides acutus, A. nobilis, Cancris subconi-cus, and Nonionella sp. (555-54,CC to 61,CC).

7. Cibicides westi dominant (555-67, 4 cm; 64 cm).8. Cribrostomoides sp. dominant with some of the

following species: Anomalinoides nobilis, Nodosaria

514

BENTHIC FORAMINIFERS

earlyMioc.

lateOlig.

mid

Eoc.

early

Eoc.

NN1

230-

NP24/25

NP16

240-

NP14

—

NP13

NP12250-

I I I

IVa

IVb

Plankton

100 0

Benthicspecies (α)

10 20 1.0

Varimax factorloadings

0.5 0 -0.5 - 1 . 0

-10,CC

-11-2,80 M 75-200

Figure 7. Hole 553A. Detail of condensed sequence shown in the center of Figure 6.

IV

spp., Lenticulina spp., Praeglobobulimina ovata, and/or Anomalinoides acutus (555-41,CC to 43,CC; 53,CC;553A-22,CC).

9. Nonionella spp. dominant with subsidiary Cancrissubconicus (555-38,CC).

10. Protelphidium sp. dominant, either alone (555-52,CC) or with Cancris subconicus and Alabamina ob-tusa (555-46,CC).

11. Trochammina sp. dominant with subsidiary El-phidium hiltermanni and Anomalinoides acutus (553A-36,CC).

Summary

It is clear that with the strong environmental controlon the distribution of species and assemblages, their usein biostratigraphy is limited. However, the following spe-cies which are confined to the late Paleocene to early Eo-cene in the Anglo-Franco-Belgian Basin (Jenkins andMurray, 1981) are also restricted to the same level at thefour sites, with the two exceptions noted:

Alabamina obtusaBolivinopsis adamsiAnomalinoides nobilis extend into middle EoceneGaudryina hiltermanni in Hole 553APulsiphonina prima, which is normally indicative of

this same interval (although it extends into the middleEocene in France; Jenkins and Murray, 1981), is con-fined to the late Paleocene to early Eocene at Sites 552to 555.

Nuttalides truempyi is confined to the Eocene.

Environmental Interpretation

The criteria for the interpretation of the Paleogeneassemblages have already been discussed.

Brackish assemblages are restricted to the late Paleo-cene-early Eocene interval. An intertidal marsh is repre-sented in 553A-36,CC by an assemblage of Trochammi-na sp. (48%), Elphidium hiltermanni (17%), and Anom-alinoides acutus (13%). The diversity is oc4, and noplanktonic forms are present.

Protelphidium sp. makes up 97% of the assemblagein 555-52,CC. The diversity is oc < 1. This is interpretedas brackish intertidal with a salinity of 10 to 25‰. Thissample contains much plant debris. Protelphidium sp.and Cancris subconicus dominate 555-46.CC, and thisis thought to represent inner shelf to outer lagoon withsalinities in the range 25 to 30‰.

Slightly brackish inner shelf (

J. W. MURRAY

0NN 20/21

PleistoceneNN19

_mioçe.ne_NN18. -early

Pliocene N N 1 3 - 1 5 c n JNN12 5 0 H

late Miocene N N 1 1

IMN9/10

NP25

Oligocene W l ^110-

early IMP22Oligocene

NP21

late Eoc. NP19/20

120H

early

EoceneNP11

NP10

-1.CC

-2,CC

3,CC

III

IV

Inferred water Plankton Benthicdepth (km) (%) species (α)

0 1 2 \ 3 0 100 0 10 20 1.0

-5,CC

-6,CC

-7,CC8,CC A

'tec"'™-2-2.3-2,CC- 3-3,80-4-2,16 3 - C C

^4-2,474-2,101

-4-3,50-4-3, 136

-4,CC

-5-1,90

- 5-2,48

-5-3,123L 5-4,12

-5,CC

-6-1,118

-6-3,4

Varimax factorloadings

0.5 0 -0.5 -1.0

Figure 8. Holes 554, 554A. Biostratigraphy based on nannozones from Backman (this volume), lithologicalunits from site chapter (this volume), and inferred water depth at the time of deposition determinedfrom the benthic foraminifers. ( = present site depth. Note the change of scale for the Paleogene.)

other bivalve mollusks. Sample 553A-13,CC with Len-ticulina sp., Cibicides, and Bolivinopsis adamsi—a di-versity of a l l , no planktonic foraminifers, and an as-sociated fauna of bryozoa, molluskan shells bored byclionid sponges, and echinoid spines—is believed to havecome from inner- to mid-shelf depths of 50-100 m.

Mid-shelf assemblages (75-150 m) are of two kinds:Those dominated by A. howelli, diversity 700 m are of early Eocene (553A-10,CC), middle Eo-cene (553A-9-6, 51 cm to 10-3, 135 cm; 552-8,CC to10,CC), late Eocene (554A-5-3, 123 cm; 5-4, 12 cm),and Oligocene (554A-4-2, 47 cm to 5-2, 48 cm) age. Allare characterized by a dominance of Nodosaria/Stilo-stomella spp., commonly with subsidiary Oridorsalisecuadorensis and sometimes with Cibicidoides spp.,Gyroidinoides spp., Globoçassidulina subglobosa, Ala-bamina wilcoxensis, Nonion havanensis, and Buliminatuxpamensis. The diversity is

BENTHIC FORAMINIFERS

Nuttallides truempyi and Osangularia spp. Some disso-lution has taken place as shown by the diversity value ofoc 11-15 and the planktonic proportion of 0-90%.

The Paleogene assemblages believed to represent thedeepest water (> 1500 m) are of Oligocene age. The domi-nant species are Nodosaria/Stilostomella spp. and Gy-roidinoides spp., with some Osangularia spp., Cibici-doides cf. kullenbergi, O. ecuadorensis, and rare B. ala-zanensis. The diversity is oc 12-24 and the planktonic tobenthic ratio is high, 94:60 to 99:1 (552A-36-4, 20 cm;552A-37-1, 12 cm; 553A-9-5, 5 cm; 553A-9-6, 19 cm).

Q-mode Varimax Factor Analysis

The 71 samples on which counts of > 100 individualswere made have been treated together for Q-mode vari-max factor analysis. The taxa were organized into 34groups as listed in Table 13. Five factors were used to de-scribe the variability. The taxonomic composition of theseis listed in Table 14, with the varimax factor scores ex-pressed as percent for each taxon.

The factors have been plotted on Figures 4 to 9, andthe distribution of each factor with respect to age andinterpreted depth of deposition is listed in Table 15. Avery clear picture is evident. Marginal marine environ-ments are represented by factors II and IV in part, butin some cases none of the factors had any significantloading. This reflects the unusual and varied composi-tion of these assemblages. The inner shelf is dominatedby factors II, III, and IV, the mid shelf by II, III, IV,and V, and the outer shelf by II and V. These environ-ments are found only in strata of late Paleocene to earlyEocene age. The slope is characterized by factors I andV. Thus, as might be expected, there is a great deal ofagreement between the subdivision into environmentsbased on a paleoecological analysis and the subdivisionusing Q-mode factor analysis. However, it should be notedthat the latter does not take into account such aspects asdiversity, planktonic to benthic ratio, or the effects ofdissolution.

FAUNAL COMPARISONS

Comparisons of faunas within the North Atlantic musttake into account both the stratigraphic age of the as-semblages and the environment of deposition. Evenwhen these conditions are satisfied problems arise be-cause of the lack of taxonomic agreement between workersin North America and Europe.

Figure 11 summarizes the depths of deposition inter-preted for each of the numbered sites. No abundancedata for the Neogene assemblages of Sites 111, 116, 117,336, 352 and the Paleogene assemblages of Sites 336 and352 are available. Also van Hinte (1976) did not give in-terpreted depths of deposition. For Site 336 he com-pared the fauna with the Barton Clay of England andthis has been interpreted by Murray and Wright (1974)as of generally inner- to middle-shelf depths, i.e., 100 m.The Oligocene fauna of Site 352 was said to indicateshallow water (van Hinte, 1976, p. 45) but was also saidto be similar to that at Site 116, which Berggren andAubert (1976b) interpret as bathyal, 800-1000 m. Van

Hinte regarded the differences in the Oligocene faunasof Sites 336 and 352 as indicative of the existence of afaunal barrier between the Atlantic and the Norwegian-Greenland Sea, i.e., the Iceland-Faeroes Ridge. This isnot a reasonable conclusion on the evidence, for if Site352 is bathyal and Site 336 shallow shelf, then two quitedissimilar environments are being compared.

Quaternary and Neogene

On the basis of depth and location the sites fall intothree groups: Iceland-Faeroes Ridge, shallow (Sites 336,352); Rockall Plateau (Sites 555, 116, 117) and OrphanKnoll (Site 111), shallow; Rockall Plateau margins, deeper(Sites 403-406, 552-554). The faunas from these regionshave been summarized in Table 16.

In the "glacial" late Pliocene and Pleistocene the fau-nas of Rockall margins are similar to those of RockallPlateau and Orphan Knoll except that some of the deeperwater indicators such as Planulina wuellerstorfi andEpistominella exigua are less common at the shallowersites. However, from the limited data in van Hinte (1976) itappears that the fauna of the Iceland-Faeroes Ridge ismuch less diverse and somewhat different in composi-tion. This reflects the modest depth and the differentwater masses influencing the area. The glacial succes-sion is characterized by the occurrence of Cassidulinateretis (recorded as C. laevigata at Sites 403-406, byMurray, 1979). However, this species was not recordedfrom Orphan Knoll by Berggren (1972). In the Norwe-gian-Greenland Sea the incoming of C. teretis (Islandi-ella teretis of van Hinte, 1976) is at ~ 5 m.y. ago in thelate Pliocene, long before the onset of glaciation in theNorth Atlantic (Schrader et al., 1976). Thus its first ap-pearance throughout the area is clearly diachronous.

The Pliocene and late Miocene assemblages are treat-ed together because they are so similar. Comparison ofthe lists in Table 16 shows that the deeper waters of Rock-all margins and the shallow waters of Rockall Plateauand Orphan Knoll have many elements in common butcertain species are rare or absent from one or the otheras a result of differing depth and water mass pref-erences, e.g., Epistominella exigua, Planulina wueller-storfi.

In the middle Miocene there are again many elementsin common, with Brizalina subaenariensis being muchmore abundant on Rockall Plateau than in the deeperwater of its margins.

In the early Miocene the depth differences betweenthe areas are modest (Fig. 11). The faunas are essentiallythe same. Siphonina tenuicarinata and Oridorsalis ecua-dorensis are both common elements. P. renzi andAnomalina alazanensis are recorded only from OrphanKnoll.

Paleogene

In the Oligocene all the sites with records were bathy-al except for Site 336 (Fig. 11). The bathyal assemblagesare composed of Cibicidoides or Heterolepa spp., No-dosaria/Stilostomella spp., Globocassidulina subglobo-sa, and Oridorsalis umbonatus or O. ecuadorensis. The

517

J. W. MURRAY

Pleist. NN20/21

"early Vl io! N N i V

NIM11

late

Mioc.100-

NN9/10

NN9/10

200-

mid

Mioc.

NN7/8

NN7/8

NN6

early

Mioc.NN3

Ila

lib

I- 1,CC

- 2,CC

- 3,CC

- 4,CC

- 5,CC

- 6,CC

- 7,CC

- 8,CC

- 9,CC

•10,CC

-12.CC

-13,CC

-14,CC

-15,CC

-16,CC

-17,CC

-18.CC

-19.CC

-20,CC

-21.CC

-22,CC

23,CC

-24,CC

-25,CC

26,CC

Inferred waterdepth (km)

0 1 | 2

Plankton(%)

0 100 0

Benthic

species (α)

10 20 30 1.0

Varima× factorloadings

0.5 0 -0 .5 - 1 . 0

10

VFigure 9. Hole 555. A. Neogene. B. Paleogene. Biostratigraphy based on nannozones from Backman (this

volume), lithological units from site chapter (this volume), and inferred water depth at the time of de-position determined from the benthic foraminifers. ( = present site depth.)

assemblage from Site 352, as already stated, is of thistype. By contrast, that at Site 336 is dominated by An-gulogerina gracilis tenuistriata and Silicosigmoilina sp.

The late Eocene assemblages of Sites 554 and 406 areessentially of Oligocene bathyal type but those of Sites111 and 116/117 differ: Site 111, Cibicidoides spp., Si-phonina wilcoxensis, Osangularia mexicana; Sites 116,117, Nuttallides truempyi, Anomalinoides grosserugo-sus. At Site 336 two distinct assemblages were recog-nized: Cibicides cf. C. tenellus; and Spiroplectamminaspectabilis and Lenticulina cultrata. Both of these areshelf assemblages.

The middle Eocene deeper water assemblages com-prise Cibicidoides spp., Nodosaria/Stilostomella spp.,Bulimina spp., O. ecuadorensis, and Osangularia sp.,sometimes with Alabamina wilcoxensis and Globocassi-dulina subglobosa. At Site 403 there is an epibathyal as-semblage of O. ecuadorensis and Gyroidinoides spp.(-150 m) and an outer shelf assemblage of Anomali-noides howelli, Cibicidoides spp., and Osangularia ex-pansa.

In the early Eocene, the only bathyal assemblages arefrom Sites 111: Cibicidoides spp., Bulimina spp., Gau-dryina hiltermanni, Nuttallides truempyi, Stilostomellaspp., and Oridorsalis ecuadorensis; and Sites 405, 406:Nodosaria/Stilostomella spp., Bulimina trigonalis andNonion havanensis (given as TV. cf. olssoni in Murray,1979). At Site 117, Berggren (1974) interpreted an as-

semblage of Gavelinella grosserugosa, Bulimina brevi-spira, Gaudryina hiltermanni, Oridorsalis ecuadorensis,Osangularis pteromophalia, and Vaginulina decorata as250-600 m (Fig. 11), but this could well be somewhatshallower (150-250 m).

The early Eocene to late Paleocene shelf assemblagesof all sites are very similar and have already been de-scribed.

In conclusion, it is evident that notwithstanding dif-ferences in taxonomy, there is a great deal of similarityin the assemblages of comparable age and environmentof deposition from the whole area.

DISCUSSIONThese results have implications for the interpretation

of the geological evolution of the area, especially therate of subsidence, the existence of faunal barriers, andthe development of water masses.

Subsidence

The pattern of subsidence is discussed in more detailin Roberts et al., this volume. Suffice it to point out that,as already established on Leg 48 (Murray, 1979), the ben-thic foraminifers show that subsidence and sedimenta-tion more or less kept pace with one another in the latePaleocene and early Eocene; all the environments are ofshallow water type, and some of the barren intervalsmay indicate emergence and even periods of eustatic

518

BENTHIC FORAMINIFERS

B£ •%

300-NP11

earlyEoc.

NP10

400-

500

?late

Paleoc.

600-

III

IVa

IVb

IVc

27-2. 105-28-1,37

Inferred water Plankton Benthicdepth (km) (%) species (α)

0 0.5 1 0 100 0 10 20

Varima× factor

loadings

1.0 0.5 0 -0 .5 -1 .0

31-1, 120-31,CC

32.CC-33.CC-34-3,60"34,CC

^36-2,3

38,CC

40.CC41,CC42,CC43,CC44.CC45.CC46.CC47.CC48.CC

52.CC53.CC54.CC55.CC56.CC57.CC58.CC59.CC60.CC61.CC

-64.CC

66.CC67-4,64

150-200

H 25-100

< 7 5

< 7 5

< 7 5

/Diss 'n

Diss'n

1

III

<

TinΦ .

• '

•II

III

iπi

π

/•H

iT

/

" A IV

Trv

• IV

M V

.''iv

Figure 9. (Continued).

sea-level change (Vail, et al., 1977). However, from thenon, the supply of sediment was inadequate to compen-sate for the rate of subsidence, so the water got progres-sively deeper. The incomplete late Eocene to early Mio-cene record, the effects of dissolution, and uncertaintyabout the ecological requirements of some Paleogenetaxa, all make it difficult to assign precise depths to envi-ronments of this period. However, all sites seem to havesubsided to around their present depths by early to mid-dle Miocene times.

Faunal Barriers

Barriers to the migration of benthic faunas may be ofmany kinds, ranging from unsuitable substrates throughsalinity and temperature gradients to emergent land mass-es. In the context of the North Atlantic, most discussionhas centered on the existence of the Greenland-Iceland-

Faeroes-Scotland Ridge as a land barrier at least duringPaleogene times (see Bott, et al., in press).

It is beyond the scope of this chapter to consider allthe arguments, geophysical, tectonic and paleontologi-cal, for and against this hypothesis. It is, however, ap-propriate to consider the evidence from the benthic for-aminifers. The key to this is a comparison of the Paleo-gene fauna of the North Atlantic with those from theNorwegian-Greenland Sea. Apart from Sites 336 and352 on the ridge itself, two other sites drilled on Leg 38yielded good early Eocene faunas. At Site 338, the earlyEocene is characterized by Lenticulina spp., Dentalinaspp., Nodosaria latejugata, and Stilostomella spinulosatogether with Spiroplectammina spectabilis, Textulariaplummerae, Bulimina cacumenata, TUrrilina brevispira,Chilostomelloides eocenica, Quadrimorphina paleocemi-ca, Melonis affinis, Pullenia quinqueloba, Anomalinoi-

519

J. W. MURRAY

Cassidulina'― teretisü < <

in 10_. inin in in in

Triloculinafrigida< <

in CM co •*in LΠ in inin in LΠ in

Buliminaalazanensis

< <in CM co *rm m m inm in in in

Laticarininapauperata

< <in CM co ^rm in m inin in in in

Ehrenbergina Ehrenberginaserrata trigona

< < < <in CM co ^ in CM co ^•m m m m i n m m i nm m m m m m m m

Uvigerina Brizalina Siphonina Globocassidulina Length ofcompressa subáenariensis tenuicarinata subglobosa succession

<< t i n c M c o ^ • i n c M c o ^ •i n i n i n i n i n m m m i n

< <in CM co rtin in m min in LΠ m

< <in CM co t̂in in in inin m in m i n m m i n m m m m

10

20-

! ε

T?

~ Hiatus

~\ Disappearance

•A Appearance

A Absent

i l

I 1

TTFigure 10. Neogene species showing restricted stratigraphic distributions.

Table 8. Q-mode varimax factors and percentage of variance accounted for, Neogene.

Hole

PleistoceneGlacial Pliocenelate Plioceneearly Pliocenelate Miocenemiddle Mioceneearly Miocene

552A

Factor

1/2/10 (4)2/4(10)1 (8)1 (8)1 (8)

Variance(%)

59-7559-6055-5954-6254-57

56

Factor

10

1—

11

(1)

553A

Variance(%)

48

5053

50-5351

50-52

Factor

1

11 (10)1

—

554

Variance

W

59-62

5956-60

5754

Factor

(4)

(1) (7)(1) (10)1/8

555

Variance(%)

14-43

43-5649-5547-51

Note: Only those factors with values 0.5 to 1.0 or -0.5 to 1.0 are listed; numbered factors in brackets onlyrarely reach these scores.

des αnomαlinoides, and Cαncris subconicus. Some ofthese species occur at Site 343 (van Hinte, 1976 in Talwani,Udintsev, et al., 1976). These forms are also present in theNorth Atlantic (Berggren, 1974; Berggren and Aubert,1976a, b; Murray, 1979; this chapter).

It could therefore be argued that on the evidence ofthe benthic foraminifers there is no reason to suspect theexistence of a land barrier between the Norwegian-Green-land Sea and the North Atlantic. But if such a barrierexisted, could the faunas on the two sides be so similar?The answer to this depends on alternative migrationroutes. The shallower water faunas of the North Atlan-tic and the Norwegian-Greenland Sea resemble those ofthe North Sea Basin, which certainly had connectionswith the North Atlantic through the English Channelfrom time to time during the Paleogene. The North Seamay also have been connected with Tethys through Eu-rope. However, the faunas of the English Channel, espe-cially its western part, are not at all like those of theNorth Sea, the Norwegian-Greenland Sea, or the Rock-all area (Murray and Wright, 1974). It therefore seemsmore likely that the North Atlantic and the Norwegian-

Greenland Sea were in communication, even if onlythrough a continuation of Rockall Trough into the Fae-roes-Shetland Basin. Furthermore, such a connectionneed not have been deep.

Water Masses

Our understanding of the development of water massesis built up on evidence of the isotopic record, sedimenttype, and distribution, existence of hiatuses, as well asthe distribution of the fauna. The distinction betweenmodern oceanic water masses is based on small differ-ences in temperature, salinity, density, etc., and this isreflected in subtle differences in faunal composition. Atthe present state of knowledge, it is not possible to de-tect such small changes in the Paleogene record and, in-deed, many would regard it as presumptuous to recog-nize it in the Neogene and Quaternary records.

Using evidence from deep-sea benthic ostracods, Ben-son (1975) suggested that cold bottom water first ap-peared in the oceans at around the Eocene/Oligoceneboundary. From oxygen isotopic evidence, Kennett andShackleton (1976) inferred the existence of a two-layer

520

BENTHIC FORAMINIFERS

Table 9. Benthic foraminiferal census data, Holes 552, 552A, Paleogene.

Species

Alabamina obtusaAlabamina wilcoxensisAlabamina sp.Anomalinoides howelliAnomalinoides nobilisAnomalinoides sp.Astrononion guadelupaeBolivinopsis adamsiBrizalina sp.Bulimina alazanensisBulimina praeinflataBulimina tuxpamensisBulimina sp.Cancris subconicusCibicides/CibicidinaCibicidoides alleniCibicidoides cf. C. bradyiCibicidoides cf. C. kullenbergiCibicidoides sp.Dentalina sp.Elphidium hiltermanniEpistominella vitreaEponides spp.Fisurina, Lagena, OolinaGaudryina hiltermanniGavellinella semicribrataGlobocassidulina subglobosaGlobulina, GuttulinaGyroidinoides spp.Lenticulina spp.Melonis sp.Nodosaria, StilostomellaNonion havanensisNonion laeveNuttalides truempyiOridorsalis ecuadorensisOridorsalis umbonatusOsangularia expansaOsangularia rugosaOsangularia sp.Pararotalia curryiPleurostomella sp.Protelphidium sp.Pullenia osloensisPullenia quinquelobaPulsiphonina primaQuinqueloculina sp.Siphonina lamarckanaSpiroplectammina spectabilisTrifarina cuneataTurrilina brevispiraUvigerina abbreviataUnidentified

Total

Number of speciesRatio planktonic to

benthic (x:l)Barren 9,CC, 13.CC

m. 1Eoc.

Hole 552

Ur )s^oo

_

—3

————2

——81

2———196

——21

—22

—591

216

——3

—————3

——2

————1

—7

25

134

390.6

U

u 10,

_18

————2

———1

4

——2

———1

———1

—961

146

——11

————3

——3

——7

———6

10

105

300.6

U

u

13

76————1

———I2

————1

—26

——5251

—2

———1

1————————1

—1

—2

—12

126

291.1

early Eocene

U

u

1

36

—

2

——

20

2—1

—

1—

2—21

—2

2——————14

———3

—18

98

280.6

•1, 7

5

oo

5

45

1—

3

——

3

92

—111

—

141

—

22

7——111

—1

12

—2

———10

116

320.8

-2,

117

8

3418

—1

217

81

2

——41224

18

—52

11————1

—6

—4

——26

150

341.9

Oligo.

-4,2

0

VOCO

1

3

3

—1

2

3131

——__

1—

281

211

—28

51

314

—4

—21

—

—1

———22

151

4016

Hole 552A

-1,

12

r~co

_

—

1

—1

7

16—

——

1—

33

1062

19

—

5

11

—4

——

1—

—4

——

111

100

2899

-1,6

5

r~CO

10

_

1—

12

———

4

——

5————

1—

1117

10—157

27

—14

—2

————

—92

——18

120

3610

1, 1

37

r~co

1

5

——

1———

——10

————

—45

—8443

—

216

——17

—1

——

2—

—11

—1

20

115

3211

middle Eocene

2,6

0

r~•CO

1

—

1——

_

——21

——1

—

6—3

—312

4—

—8

—1

——1

—

1———49

49

242.7

•3, 6

7

r~co

_

16

_

———

—18—

2

——1

———61

——11—8127

10—12

———1

—1

————

1————13

102

280.2

1, 6

2

ooCO

_

8

—————

112—

———3———1——1——3119———1———3—2——2—

—————5

53

170.09

•3, 2

7

ooCO

_

40

———————10

—

1———

8———

12

—1

——

111

—9

——————

1——

2——

2—

4————11

104

260

Uüoo"CO

1

—————————

3

———————

2——————

2———————————

1———————————

9

50

Note: Dash = absent in assemblage count.

ocean, with the lower layer (psychrosphere) having atemperature of less than 10°C, at the Eocene/Oligoceneboundary. This stratification was attributed to the pro-duction of extensive sea ice around Antarctica which, inturn, promoted cold bottom-water formation and more

vigorous bottom circulation. The effect of these changes isevident in changes in sediment distribution in the oceansat this period (see, for example, Moore, et al , 1978).Recently, Corliss (1981) has reviewed the effect of thiscooling on benthic foraminifers. His data for Site 277 in

521

Tkble 10. Benthic foraminiferal census data, Hole 553A, Paleogene.

Oligo. middle Eocene early Eocene

Species

— — — — — 24e

_ _ _ _ 4

3 — —15 — —

— — 12

1

_ _ — — — — 3

Alabamina obtusaAlabamina wilcoxensisAlabamina sp.Anomalinoides acutusAnomalinoides howeliiAnomalinoides nobilisAnomalinoides sp.Astrononion guadelupaeBolivinopsis adamsiBrizalina sp.Bulimina alazanensisBulimina praeinflataBulimina trigonalisBulimina tuxpamensisBulimina sp.Cancris subconicusCibicides westiCibicides/CibicidinaCibicidoides alleniCibicidoides cf. C. bradyiCibicidoides cf. C. kullenbergiCibicidoides spp.Cribrostomoides sp.Cyclammina sp.Dentalina sp.Elphidium hiltermanniEpistominella vitreaEponides umbonatusEponides spp.Fissurina, Lagena, OolinaFlorilus elongatumGaudryina hiltermanniGavelinella semicribrataGlobocassidulina subglobosaGlobulina, GuttulinaGyroidinoides spp.Lenticulina spp.Loxostomoides applinaeMelonis sp.

Nodosaria, SlilostomellaNonion havanensisNonion laeveNonionella spp.Nuttallides truempyiOridorsalis ecuadorensisOridorsalis umbonatusOsangularia rugosaOsangularia sp.Pararotalia curryiPleurostomella sp.Praeglobobulimina ovataProtelphidium sp.Pullenia bulloidesPullenia osloensisPullenia quinquelobaPullenia sp.Pulsiphonina primaQuinqueloculina spp.Siphonina lamarckanaSpiroplectammina spectabilisTrifarina cuneataTrochammina spUnidentified

Total

Number of speciesRatio planktonic to

benthic (x:l)Barren 15,CC, 17,CC, 26,CC, 2g,CC, 31.CC, 33.CC, 34,CC.

2 — - - - —

— 12 — —

— — — — 242

— — 2 3 2

2 -5 —

— — 14

20 — —— — 2

— — — — 62 — —

— — 2

— — — 26

717 — —2 — -

— — 6

10 — — — — — — — —_ _ _ _ 1 _ _ _ 49

— — — 1

— — — — — 5— — — 37

1 — — —- — — 1- — — 2

201

3

BENTHIC FORAMINIFERS

Table 11. Benthic foraminiferal census data, Hole 554A, Paleogene.

Species

Alabamina sp.Anomalinoides howelliAnomalinoides nobilisAnomalinoides sp.Astrononion guadelupaeBolivinopsis adamsiBrizalina subaenariensis/?r/V/7//rt/7 rf R thnlmnnni‰ßt t4,utlflu VI. U. lllUllllUlltlt

Brizalina sp.Bulimina alazanensisBulimina trigonalisBulimina tuxpamensisBulimina sp.Ceratolamarckina sp.Cibicides, CibicidinaCibicidoides alleniCibicidoides bradyiCibicidoides kullenbergiCibicidoides spp.Eggerella bradyiEpistominella exiguaEponides umbonatusFissurina, Lagena, OolinaGavelinella semicribrataGlobocassidulina subglobosaGlobulina, GuttulinaGyroidinoides spp.Karreriela bradyiMelonis sp.Nodosaria spp./Stilostomella spp.Nonion havanensisNonion sp.Nonionella sp.Oridorsalis ecuadorensisOridorsalis umbonatusOsangularia rugosaOsangularia sp.Pararotalia calvezaePleurostomella sp.Pullenia bulloidesPullenia osloensisPullenia quinquelobaPulsiphonina primaSpiroplectammina spectabilisTrifarina cuneataUvigerina compressaUvigerina sp.Unidentified

Total

Number of speciesRatio planktonic to

benthic (x:l)

r~

Table 12. Benthic foraminiferal census data, Site 555, Paleogene.

early Eocene ?late Paleocene

εio O w

© r~ r*J •

BENTHIC FORAMINIFERS

Tkble 13. Paleogene species groups used in the varimax factor analysis.

1. Alabamina obtusa2. Alabamina spp.3. Anomalinoides acutus

A. spp., A. howelliCibicides, Cibicidina

4. Gavelinella semicribrata5. Anomalinoides nobilis6. Bulimina tuxpamensis1. Bulimina & Brizalina8. Cancris subconicus9. Cibicidoides alleni

10. Cibicidoides spp.11. Cribrostomoides/Trochammina12. Elphidium hiltermanni13. Epistominella vitrea14. Eponides spp.15. Gaudryina spp.16. Globocassidulina subglobosa

17. Gyroidinoides spp.18. Lagena, Fissurina, Globulina, Guttulina19. Lenticulina spp.20. Melonis spp.21. Nodosaria/Stilostomella22. Nonion laeve23. Nonion havanensis24. Nonionella/Florilus25. Nuttatlides truempyi26. Oridorsalis spp.27. Osangularia spp.28. Pleurostomella spp.29. Protelphidium spp.30. Pullenia spp.31. Pulsiphonina prima32. Spiroplectammina spectabilis33. Uvigerina spp.34. Praeglobobulimina ovata

Table 14. Paleogene assemblages represented byvarimax factors (factor scores × 100).

Varimax factor I

62.9 Nodosaria/Stilostomella spp.39.5 Gyroidinoides spp.37.5 Pullenia spp.33.9 Oridorsalis spp.24.0 Cibicidoides spp.

Varimax factor II

96.9 Anomalinoides acutus, A. howelli,Cibicides, Cibicidina

Varimax factor III

89.7 Praeglobobulimina ovata22.4 Nonionella sp.21.3 Cancris subconicus

Varimax factor IV

-64.2-44.8-33.4-24.1-20.1-19.2

Lenticulina spp.Cribrostomoides/TrochamminaAnomalinoides nobilisNodosaria/StilostomellaCibicidoides alleniAlabamina obtusa

Varimax factor V

-54.8-51.8-43.8-29.6-25.5-18.1

Alabamina sp.Osangularia sp.Bulimina tuxpamensisNonion havanensisCibicidoides spp.Nuttallides truempyi

As much of the Norwegian Sea Paleogene record hasbeen badly affected by dissolution, it is tempting to in-voke the introduction of cool Norwegian Sea water intothe area as the cause of both the vigorous bottom circu-lation and the dissolution. However, the influence ofAntarctica cannot be ignored, for this played a majorrole in Paleogene circulation of the South Atlantic, andvan Andel et al. (1977) describe the change in the deepwater circulation there as from a preglacial to a pro-toglacial pattern.

Faunal similarities within comparable depth zones showthat in shelf seas there was faunal communication be-tween such widely separated areas as Rockall Plateau,

Vòring Plateau, and the North Sea during the early Eo-cene. In deeper water there was communication betweenOrphan Knoll and the southern margin of Rockall Pla-teau (Sites 405, 406) in early and middle Eocene times,and between Rockall margin (Sites 552, 553, 554), Hat-ton-Rockall Basin (Sites 116, 117), and the Iceland-Faeroes Ridge (Site 352) in Oligocene times.

Recent reviews of the Cenozoic development of deepcirculation in the North Atlantic have favored an inputof cold water from the Norwegian Sea by late Eocene toearly Oligocene times (Berggren and Hollister, 1974; Berg-gren and Schnitker, in press; Miller and Tucholke, inpress). The interpreted existence of NADW assemblages(based on the Q-mode varimax factor analyses from atleast early Miocene time is consistent with, and lendssome support to, this interpretation.

ACKNOWLEDGMENTSI am grateful to Ms. J. Weston for making available her data on

the modern benthic Foraminifera of the Northeast Atlantic, for carry-ing out much of the computing, and for discussion. Problems of tax-onomy were discussed with Dr. R. G. Douglas and Ms. F. Woodruff ofthe University of Southern California; Drs. B. Corliss, P. Belanger,and K. Miller of Woods Hole Oceanograhic Institution; Dr. W. Poagof the U.S. Geological Survey; and Dr. D. Schnitker of the Universityof Maine. Dr. R. Riding, University College of Wales, Cardiff, andDr. G. Elliott, British Museum (Natural History) identified the dasy-clad alga. The author is grateful for financial support from the Natu-ral Environment Research Council to make visits to research institutesin the United States. Ms. G. Wright typed the manuscript and Mr. J.Jones prepared the photographs in Plates 1 to 4 from negatives takenby the author. Dr. J. R. Haynes, University College of Wales, Aberyst-wyth, and Dr. D. Schnitker, University of Maine, kindly commentedon the manuscript. The reviewers were Dr. W. Poag, U.S.G.S., and Dr.B. Corliss, Woods Hole.

REFERENCES

Alvinierie, J., Caralp, H., Latouche, C , Moyes, J., and Vigneaux,M., 1978. Apport à la connaissance de la paléohydrologie de 1'At-lantique nord-oriental pendant le Quaternaire terminal. Oceanol.Ada, 1:87-98.

Belanger, P. E., and Streeter, S. S., 1980. Distribution and ecology ofbenthic foraminifera in the Norwegian-Greenland Sea. Mar. Mi-cropaleontol., 5:401-428.

Benson, R. H., 1975. The origin of the psychrosphere as recorded inchanges of deep-sea ostracod assemblages. Lethaia, 8:69-83

Berger, W. H. and Winterer, E. L., 1974. Plate stratigraphy and thefluctuating carbonate line. In Hsü, K. J., and Jenkyns, H. (Eds.),Pelagic Sediments on Land and Under the Sea. Spec. Publ. Int.Assoc. Sedimentol., 1:11-48.

Berggren, W. A., 1974. Late Paleocene-Early Eocene benthonic fora-miniferal biostratigraphy and paleoecology of Rockall Bank. Mi-cropaleontology, 20:426-448.

Berggren, W. A., and Aubert, J., 1976a. Eocene benthonic foraminif-eral biostratigraphy and paleobathymetry of Orphan Knoll (Labra-dor Sea). Micropaleontology, 22:327-346.

, 1976b. Late Paleogene (Late Eocene and Oligocene) ben-thonic foraminiferal biostratigraphy and paleobathymetry of RockallBank and Hatton-Rockall Basin. Micropaleontology, 22:307-326.

Berggren, W. A., and Hollister, C. D., 1974. Paleogeography, pa-leobiography and the history of circulation of the Atlantic Ocean.Soc. Econ. Paleontol. Mineral. Mem., 20:126-186.

Berggren, W. A., and Schnitker, D., in press. Cenozoic marine envi-ronments in the North Atlantic and Norwegian-Greenland Sea.

Boltovskoy, E., and Wright, R., 1976. Recent Foraminifera: TheHague (Junk), pp. 1-515.

Bott, M., Talwani, M., Thiede, J.,and Saxov, S. (Ed.), in press. Thestructure and development of the Greenland-Scotland Ridge: Newmethods and concepts. Proc. NATO Advanced Research Institute.

Cooper, L. H. N., 1955. Deep water movements in the North Atlanticas a link between climatic changes around Iceland and biological

525

J. W. MURRAY

productivity of the English Channel and Celtic Sea. J. Mar. Res.,4:347-362.

Corliss, B. H., 1981. Deep-sea benthonic foraminiferal fauna turnovernear the Eocene/Oligocene boundary. Mar. Micropaleontol., 6:

*367-384.Culver, S. J., and Buzas, M. A., 1980. Distribution of Recent benthic

foraminifera off the North American Atlantic coast. SmithsonianContrib. Mar. sci., 6:1-512.

Douglas, R., and Woodruff, R, 1981. Deep sea benthic foraminifera.In: Emiliani, C. (Ed.), The Sea (Vol. 7): New York (Plenum),1233-1327.

Haake, F. W., Coulbourn, W. T., and Berger, W. H., 1982. BenthicForaminifera: depth distribution and redeposition. In von Rad, V.,Hinz, K., Sarnthein, M., and Seibold, E. (Eds.), Geology of theNorthwest African Continental Margin: Berlin (Springer Verlag),pp. 632-657.

Jenkins, D. G., and Murray, J. W., 1981. Stratigraphical Atlas of Fos-sil Foraminifera: Chichester (British Micropalaeontological Series,and Ellis Horwod), pp. 1-310.

Kennett, J. P., and Shackleton, N. J., 1976. Oxygen isotope evidencefor the development of the psychrosphere 38 Myr ago. Nature,260:513-515.

Lagoe, M. B., 1977. Recent benthic foraminifera from the CentralArctic Ocean. J. Foram. Res., 7:106-129.

Laughton, A. S., Berggren, W. A., et al., 1972. Init. Repts. DSDP,12: Washington (U.S. Govt. Printing Office).

Lohman, G. P., 1978. Abyssal benthonic Foraminifera as hydrograph-ic indicators in the western South Atlantic Ocean. J. Foram. Res.,8:6-34.

Lonsdale, P., and Hollister, C. D., 1979. A near-bottom traverse ofRockall Trough hydrographic and geological inferences. Oceanol.Acta, 2:11-105.

Lutze, G. F. 1980. Depth distribution of benthic foraminifera on thecontinental margin off NW Africa. Meteor Forsch.-Ergebnisse,32:31-80.

Miller, K. G., and Lohmann, G. P., 1982. Environmental distributionof Recent benthic foraminifera in the northeast United States con-tinental slope. Geol. Soc. Am. Bull., 93:200-206.

Miller, K. G., and Tucholke, B. C , in press. Development of Cenozoicabyssal circulation south of the Greenland-Scotland Ridge. Proc.NATO Advanced Research Institute.

Moore, T. C , Jr., van Andel, Tj. H., Sancetta, C , and Pisias, N.,1978. Cenozoic hiatuses in pelagic sediments. Micropaleontology,24:113-138.

Murray, J. W, 1973. Distribution and Ecology of Living Benthic Fora-miniferids: London (Heinemann).

, 1976a. Comparative studies of living and dead benthic fo-raminiferal distributions. In Hedley, R. H., and Adams, C. G.,(Eds.), Foraminifera, 2:45-109.

., 1976b. A method of determining proximity of marginal seas toan ocean. Mar. Geol., 22:103-109.

., 1979. Cenozoic biostratigraphy and paleoecology of Sites403 to 406 based on the foraminifers. In Montadert, L., Roberts,D. G., et al., Init. Repts. DSDP, 48: Washington (U.S. Govt. PrintingOffice), 415-430.

Murray, J. W., and Wright, C. A., 1974. Palaeogene Foraminiferidaand palaeoecology, Hampshire and Paris Basins and the EnglishChannel. Spec. Pap. Palaeontol., 14:1-129.

Phleger, F. B., 1952. Foraminiferal distribution in some sediment sam-ples from the Canadian and Greenland Arctic. Contrib. CushmanFound. Foram. Res., 3:80-89.

Phleger, F.B., Parker, F. L., and Peirson, J. R, 1953. North AtlanticForaminifera. Rept. Swed. Deep-Sea Exped., 7:1-122.

Schafer, C. T., and Cole, F. E., 1982. Living benthic foraminifera dis-tributions on the continental slope and rise east of Newfoundland,Canada. Geol. Soc. Am. Bull., 93:207-217.

Schnitker, D., 1974. West Atlantic abyssal circulation during the past120,000 years. Nature, 248:385-387.

, 1979a. Cenozoic deep water benthic foraminifers, Bay ofBiscay. In Montadert, L., Roberts, D. G., et al., Init. Repts. DSDP,48: Washington (U.S. Govt. Printing Office), 377-413.

., 1979b. The deep waters of the western North Atlantic dur-ing the past 24,000 years, and the re-initiation of the Western Bound-ary Undercurrent. Mar. Micropaleontol., 4:265-280.

_, 1980. Quaternary deep-sea benthic foraminifers and bot-tom water masses. Ann. Rev. Earth Planet. Sci., 8:343-370.

Schrader, H. J., Bjòrklund, K., Manum, S., Martini, E., and van Hinte,J., 1979. Cenozoic biostratigraphy, physical stratigraphy, and pale-oceanography in the Norwegian-Greenland Sea, DSDP Leg 38 pa-leontological synthesis. In Talwani, M., Udintsev, G., et al., Init.Repts. DSDP, 38: Washington (U.S. Govt. Printing Office),1197-1211.

Sen Gupta, B. K., Lee, R. R, and May, M. S., 1981. Upwelling and anunusual assemblage of benthic foraminifera on the northern Flori-da continental slope. /. Paleontol., 55:853-857.

Streeter, S. S., 1973. Bottom water and benthonic Foraminifera in theNorth Atlantic—Glacial-Interglacial cycles. Quat. Res., 3:131-141.

Streeter, S. S., and Lavery, S. A., 1982. Holocene and latest glacialbenthic foraminifera from the slope and rise off eastern North Amer-ica. Geol. Soc. Am. Bull., 93:190-199.

Talwani, M., Udintsev, G., et al., 1976. Init. Repts. DSDP, 38: Wash-ington (U.S. Govt. Printing Office).

Vail, P. R., Mitchum, R. M., and Thompson, S., 1977. Seismic stra-tigraphy and global changes of sea level, Pt. 4: global cycles of rel-ative changes of sea level. Am. Assoc. Pet. Geol Mem., 26:83-97.

van Andel, Tj. H., Thiede, J., Sclater, J. G., and Hay, W. W, 1977.Depositional history of the South Atlantic Ocean during the last125 million years. J. Geol., 85:651-698.

van Hinte, J. E., 1976. Sites 336 and 352t In Talwani, M., Udintsev,G., et al., Init. Repts. DSDP, 38: Washington (U.S. Govt. PrintingOffice), 23-116.

Vergnaud Grazzini, C , Müller, C , Pierre, C , Létolle, R., and Pey-pouquet, J. P., 1979. Stable isotopes and Tertiary paleontologicalpaleoceanography in the Northeast Atlantic. In Montadert, L., Rob-erts, D. G., et al., Init. Repts. DSP, 48: Washington (U.S. Govt.Printing Office), 475-487.

Williams, C. B., 1964. Patterns in the Balance of Nature: London(Academic Press).

Worthington, L. V., 1970. The Norwegian Sea as a Mediterranean ba-sin. Deep-Sea Res., 17:77-84.

, 1976. On the North Atlantic circulation. John HopkinsOceanogr. Sen, 6:1-110.

Worthington, L. V., and Wright, W. R., 1970. North Atlantic OceanAtlas of potential temperature, and salinity in the deep water, in-cluding temperature, salinity and oxygen profiles from the "ErikaDan" cruise of 1962. Woods Hole Oceanogr. Inst. Atlas Ser. (Vol.2).

Date of Acceptance: May 24, 1983

526

Table 15. Distribution of Q-mode varimax factors in the Paleogene samples.

BENTHIC FORAMINIFERS

Marginal

Inner shelf

Mid shelf

Outer shelf

Slope > 7 0 0 m

Slope > 1500 m

latePaleocene

earlyEocene

middleEocene

lateEocene Oligocene

Sites

552 553 554 555 552 553 554 555 552 553 554 555 552 553 554 555 552 553 554 555

IIIIIIV

IIIIIIV

II

IIIVII

IVIIIIVVIIVIV

IIIIIIV

Figure 11. Depths of deposition (in meters) of North Atlantic DSDP sites based on analysis of benthic fora-miniferal assemblages. (1. Berggren, Laughton et al., 1972; Berggren and Alibert, 1976a; 2. Murray,1979; 3. this paper; 4. Berggren, Laughton, et al., 1972;