Late Minoan Jar ca. 1450-1400 bce

Palaeo-oceanography and biogeography in the Tremadoc
(Ordovician) Iapetus Ocean and the origin of the chemostratigraphy
of Dictyonema flabelliforme black shales


* Marine Sciences Group, Department of Paleontology, University of California, Berkeley, California 94720, U.S.A.
** Isotope and Nuclear Chemistry Division, Los Alamos National Laboratory, Los Alamos, New Mexico, 87545 U.S.A.

(Received 21 July 1987; accepted 24 April 1988, Geological Magazine. v. 126, p. 19-27 (1989) Revised for WWW April 1997

High concentrations of vanadium, molybdenum, uranium, arsenic, antimony with low concentrations of manganese, iron and cobalt heretofore restricted to Dictyonema flabelliforme bearing Tremadoc black shales in Balto-Scandia, have been found in coeval black shales in the Saint John, New Brunswick area. Prior palaeogeographic reconstructions place these areas about 400 km. apart in high southern latitudes in the Iapetus Ocean, with New Brunswick in proximity to Avalonia (southeastern Newfoundland). These geochemical similarities are not found in coeval Tremadoc black shales of Bolivia, New York, Quebec, Wales, and Belgium. Palaeo-oceanographic reconstructions of Iapetus support the proximity of Balto-Scandia and the Saint John area during the early Tremadoc and Gee's (1981) suggestion that the signature is a feature of eastern Iapetus. Furthermore, first-order modelling of the major surface currents and related primary productivity in the Tremadoc Iapetus Ocean explain the apparent wide latitudinal range of D. flabelliforme (Fortey, 1984) and the anomalous trace metal content of certain black shales of that time. Variations in the elemental content of these black shales is produced by oceanographic and geologic conditions unique to the geographic site. The distinctive Balto-Scandic geochemical signature resulted from the coincidence of anoxic waters transgressing the shelf at latitudes of high organic productivity at the polar Ekman planetary divergence. This produces the conditions for concentrations of V, U. and Mo in the shales. Metal enriched anoxic bottom waters produced by leaching of volcanics or through hydrothermal activity may be the source of the other enhanced signature elements such as As and Sb. The absence of this geochemical signature in younger non-D. flabelliforme Tremadoc and later black shales in Balto-Scandia and other areas suggests that the closing of Iapetus moved the depositional sites into less productive oceanic areas.

1. Anomalous chemistry of the Tremadoc BaltoScandian black shales

Unusually high concentrations of vanadium, molybdenum, and uranium from the Alum Shale of Sweden (Mid-Cambrian to Early Ordovician) and the Tremadoc Dictyonema Shales in the Balto-Scandian region have been cited by many authors (Goldschmidt, 1954; Vine & Tourtelot, 1970; Armands, 1972; Bjorlykke, 1974a; Andersson, Dahlman & Gee, 1982; and Sundblad & Gee (1985). Gee (1981) noted the faunal correspondence of Dictyonema flabelliforme {= Rhabdinopora flabelliforme. For this paper we have retained the genus name Dictyonema as it is more familiar to non-graptolite specialists.} with rocks with the unique chemical signature in Scandinavia and suggested a possible linkage that was characteristic of the 'eastern' side of Iapetus in Tremadoc times. Berry et al. (1986), using neutron activation analysis (Minor et al. 1981), reported additions to the geochemical signature as (1) antimony and arsenic also showed high concentrations and (2) manganese, iron and cobalt showed low concentrations in the Oslo region D. pabelliforme shales. Berry et al. (1986) also investigated D.flabelliforme shales from Wales; Levis, Quebec; Schaghticoke, New York; and from the Bolivian Cordillera, but could not find the characteristic geochemical signature of coeval Balto-Scandian shales. However, neutron activation analyses of additional samples of Tremadoc shales, reported here, from Sweden, Estonia, Belgium, and Canada confirmed the expanded list of signature elements for Balto-Scandia. We also found that D. flabelliformebearing shales from New Brunswick, Canada have the Balto-Scandian geochemical signature; but that the coeval Belgium and other Canadian samples do not. Table I gives the comparison among Balto-Scandian, New Brunswick, and other Tremadoc shales in the D. flabelliforme Zone for the signature elements. The complete signature in marine black shales, consisting of both high and low elemental concentrations, to our knowledge, is restricted to Upper Cambrian (Swedish Alum Shale) and the Tremadoc rocks reported here. Although high V, Mo, and U is not uncommon in black shales (Wedepohl, 1964), the corresponding high signature of As and Sb and the low signature of Mn, Fe, and Co is not found, for example, in the Precambrian (1800 my) black schist of Finland (Peltola, 1968); the Permian Kupfersheifer of Germany (Wedepohl, 1964); or the Cretaceous sapropels of the South Atlantic (Degens et al. 1986; Brumsack, 1986). Carbon and sulfur analyses were not reported, as Berner & Raiswell (1983) and Raiswell & Berner (1986, 1987) have found that such analyses are unreliable and non-reproducible for 'weathered' samples, particularly for older Lower Palaeozoic outcrop and museum specimens of the type analysed here. Unfortunately, reliable C and S analyses can be made only on unweathered core material.

Table 1. Signature elements, Tremadoc black shales.

Abundances in parts per million
Balto-ScandiaNew BrunswickOther Tremedoc*
(47 samples)(5 samples)(44 samples)
"Organic" Signature
V472150190626051100410134044 150 70 320
Zn287507137800NDNDND9 103 7281
Mo4514046005 248 496 8 0.1 25
U4763 7440521 12 2444 4 2 11
High Signature: 'volcanic' and other sources
As4739 5815 23 20 2944 13 2 42
Se31 3.36.25 2.9 1.0 6.6
Br 32 6.20.936.15 14.3 4.9 18.22 2.8 2.72.8
Sb4714 1335 5.1 4.5 5.634 1.4 0.3 4.6
Ba473750160 5020052160 1730 262044 860 160 2180
Ta 471.20.6 1.85 1.3 0.8 1.744 0.8 0.3 1.8
Low Signature: redox indicators
Mn47 15025 7805 36 32 4144 370 48 2460
Fe47 227004600 414005 15000 12100 1620044 40000 10360 99790
Co4712.2 0.431.758.8 0.9 12.144 16.6 1.3 54.6
* Other Tremadoc locations include Wales (15); Canada including Levis, Matane and Gaspe (16); New York (4); Belgium (3) and Bolivia
"n"' values list the number of samples of the total analysed that were above the detection limits for NAA.
ND. Not detected by NAA.

2. Palaeogeographical reconstructions

The palaeogeography of the Lower Ordovician may give a clue to the relationship among sites with and without the unusual geochemical signature. Cocks & Fortey (1982) assigned palaeolatitudes and longitudes to their palaeogeographic reconstruction of the Arenig Iapetus Ocean making predictions of climatic and oceanographic conditions possible. Reversing their closing rate of 2 cm per year, we suggest (Fig. 1) a reconstruction of Iapetus for Tremadoc times (Whittington et al. 1984) essentially showing the maximum extent of the Iapetus Ocean. This reconstruction is corroborated on a global scale by Scotese (1986). The Cocks & Fortey (1982) reconstruction shows considerable longitudinal distance (about 4000 km) between Balto-Scandia and New Brunswick and places them on different continental blocks separated by the Tornquist Sea. That reconstruction places southern New Brunswick with Nova Scotian Maguma Terrane so that during closure of Iapetus the present orientation, west to east, of New Brunswick-Nova Scotia, Newfoundland, Ireland, and England is maintained.

However, samples from Belgium and Wales on the same block as southen New Brunswick, in their reconstruction, do not show the geochemical signature. An alternative reconstruction, based on the geochemical affnities following Gee's (1981) suggestion, would place New Brunswick on the same block as Balto-Scandia during the Tremadoc on the eastern side of Iapetus. Then, during closure of Iapetus, southern New Brunswick was transformed to align with Nova Scotia. If the transform occurred during or after the closer of the Tornquist Sea, then the longitudinal movement of southern New Brunswick along the transform fault would be reduced. Figure 1 shows both potential positions for New Brunswick during the Tremadoc.

3. Palaeo-oceanography

3.a. Circulation model

Figure 1. shows a major southern hemispheric ocean with an eastern meridional boundary and major landmasses near the Equator and 60° S. This permits use of the Sverbrup-Ekman model (Stommel, 1957) to reconstruct the major circulation patterns. Parrish (1982) has used a similar approach emphasizing the influence of Ekman transport on the degree and position of upwelling. Assuming a similarity with the present planetary wind patterns, or polar and equatorial zonal easterly and temperate zonal westerly winds; Ekman transport vectors in the oceanic upper mixed layer driven cum sol to the zonal winds would be directed (1) polarward in Polar and Equatorial climatic zones and (2) equatorward in the Temperate zone. The reversal of the direction of the Ekman transport vectors at the climatic boundaries coupled with the reversal of the horizontal Coriolis deflection at the Equator would produce planetary divergences at the Equator and in the vicinity of 60° S and a planetary convergence at 30° S. The addition of meridonal or north-south land boundaries or submerged barriers would interrupt the oceanic flow and produce (a) a low pressure oceanic gyre centred on a planetary divergence and (b) a high pressure oceanic gyre centred on a planetary convergence.

3.a.1. Application to the Tremadoc lapetus Ocean

For the southern hemisphere location of Iapetus in the Tremadoc, the major oceanic circulation would be clockwise about low pressure gyres and counterclockwise about high pressure gyres. The primary circulation pattern for the palaeogeographic reconstruction given in Figure 1 would be dominated by counter-clockwise circulation about the oceanic high centred at 30° S. Along the southern shores of Iapetus, the eastward flowing component of counter-clockwise gyre would parallel the Eur-African coast, then turn northward at the meridonal boundary of BaltoScandia. The presence of the Tornquist Sea would show little influence on the Sverdrup-Ekman model circulation because the sea seemingly was relatively small and shallow embayment compared to Iapetus. The apparent large areal extent of the Tornquist Sea as shown in Figure 1 is caused by the arealdistortion at high latitudes of the Mercator projection. The circulation in such an embayment at these latitudes would be clockwise and complement the low pressure circulation at the 60° S planetary divergence. Fig. 1 also shows that some exchange of water and potentially fauna could occur at the northern exit of Tornquist. At the meridional boundary of BaltoScandia, the flow would deflect northward toward the Equator, parallel to the boundary. The north-south barrier does not necessarily have to be a landmass. A submerged shelf or bank is sufficient to deflect the flow in the Sverdrup Ekman model. Near the Equator, the counter-clockwise circulation about the mid-latitude high would mesh with the clockwise circulation about the equatorial low pressure system and the flow would deflect to the west, parallel to the Laurentian landmass as the South Equatorial Current. At the Equator, an Equatorial Counter Current might develop which would flow eastward parallel to a northern hemispheric Laurentian landmass. The presence of islands or a continental landmass along the Equator might interrupt peripheral planetary circulation or produce monsoonal conditions which could seasonally reverse nearshore current patterns.

This interpretation differs from that of the nearest in time given by Parrish (Franconian {Late Cambrian}: 1982, p. 759) where Balto-Scandia is west, rather than east and south, of Laurentia and both blocks are at more middle latitude positions than Cocks & Fortey (1982) assign to the Arenig. Extrapolaton of Parrish's Late Cambrian palaeogeography and interpretations of atmospheric circulation and zones of upwelling into the Tremadoc would be inconsistent with the distribution of Dictyomena flabelliforme and patterns of upwelling described for Balto-Scandia by Lindstrom & Vortisch (1983).

3.b. Nutrient distribution

Primary productivity would be enhanced in the vicinity of 0° and 60° S due to upwelling (Ekman pumping) or P. N. and Si-rich waters along the major planetary Ekman divergence zone in a pattern described by LaFond & LaFond (1971) for the modern ocean. In either potential position of New Brunswick (Fig. 1), both Balto-Scandia and New Brunswick are within the major zone of enhanced primary productivity, in the vicinity of 60°. Upwelling also would be produced by entrainment and off shore advection because of the equatorward flow at the eastern meridional boundary (Stommel, 1957). This type of upwelling would be likely on the north and western margins of a submerged Balto-Scandian block (Fig. 1). At midlatitudes, the eastern shore of Iapetus would have seasonal spring-summer upwelling caused by Ekman Transport produced by seasonal non-zonal winds, as seen in the modern ocean (Wooster & Reid, 1963).

Such seasonal upwelling, occurring during maximum insolation, would extend the enhanced productivity northward from the main divergence at 60° S. These oceanographic mechanisms operating at the geographic positions shown in Figure 1 support Lindstrom & Vortisch's (1983) proposal of increased productivity for the Balto-Scandian shelf in the Early Ordovician.

3.c. Apparent eurygraphic distribution of D. flabelliforme

Cooper (1979) plotted the geographic range of D. flabelliforme in the Tremadoc as a tropical to subtropical fauna, symmetrically arranged about the Equator. Erdtmann (1982, 1984) gave this fauna a wider latitudinal range, but restricted it essentially to the southern hemisphere. The reconstruction given in Figure 1 shows a distribution in the southern hemisphere of at least 60° latitude and at least several thousand kilometres in longitude This map does not include D. flabelliforme from Siberia or Argentina-Bolivia or possible flabelliforme-group taxa from Algeria (Legend, 1974). Such distributions imply a eurythermal tolerance because the fossil evidence is placed from the tropics to near the palaeogeographic pole in Tremadoc times. However, the widespread occurrence of black shales, indicative of anoxic bottom waters on the shelf; as well as high stands of sea level, suggest a warmer climate in the Tremadoc than during glacial-interglacial intervals. Thus, during warm climates, the influence of the midlatitudinal gyre would expand polarward.

In the modern ocean, the warm water cosmopolite faunas (McGowan, 1972, p. 15) have a range of 40° north and south of the Equator. The modern cosmopolites are associated with the major oceanic gyre driven by the mid-latitude oceanic high pressure, which is described by the Ekman model used above. By analogy, D.flabelliforme could have been part of a warm water cosmopolite fauna. Its wide latitudinal range in the Tremadoc may be attributed to relative expansion of the influence of major mid-latitude circulation gyre during a warm climatic interval, compared to modern cooler intervals. This approach harmonizes the apparent disparate geographical reconstructions of Cooper (1979) and Erdtmann (1982) and provides oceanographic substantiation of Fortey's (1984) view that D. flabelliforme was a deep-water generally pandemic species which expanded shelfward in the Tremadoc transgression. The paucity of D. flabelliforme and other planktic graptolitic faunas in association with undoubted shallow water faunas suggests an open ocean environment influenced by the major ocean currents only. Coeval trilobite faunas from southeast New Brunswick belong to the AcadoBaltic faunal province (Landing, Taylor & Erdtmann (1978). These trilobite faunal affinities are consistent with our reconstruction. Any occurrence of D. flabelliforme with shallow water or near shore fossils might be due to the spinning off of warm-core eddies, containing planktonic graptolites, from the expanded central oceanic gyre onto the shelf areas. Yamamota & Nishizawa (1986) and Conte, Bishop & Backus (1986) describe modern examples of the biological consequences of incursion of warm-core eddies onto shelf seas. This would occur most frequently at high latitudes on the western side of the ocean as the main current moved away from the meridional boundary landmass near the planetary divergence (Wiebe & McDougall, 1986). Thus the oceanographic currents depicted using the Sverdrup-Ekman model and the distribution of planktic D. flabelliforme fossils in black shales are consistent with the palaeogeographic reconstruction of Figure I, without resort to additional geologic evidence from shallow water faunas.

4. Deposition of D. flabelliforme shales

4.a. Source mechanisms for signature elements

4.a.1. Redox related elements

The low signature elements Mn and Fe, which are redox sensitive, may be indicative of the anoxicity of the transgressing waters and increased anoxicity in organic rich sediments (Wilde et al. 1984). Low Mn is thermodynamically easy to explain in anoxic conditions in the absence of carbonates (Stumm & Morgan, 1970; Froelich et al. 1979) as the most common Mn minerals are oxides and carbonates. However, Fe should be relatively high, associated with common iron sulfides such as pyrite. A possible explanation is that at this level of anoxicity, ammonia complexes of Fe, which are very soluble, impede precipitation of iron sulphides. This suggests that the planktonic organic matter, the anoxic bottom water (Wilde & Berry, 1986; Wilde, 1987) or both are rich in reduced nitrogen compounds. A better explanation of the low Mn and particularly the low Fe signature awaits the development of an acceptable geochemical model of anoxic zonation in the ocean and bottom sediments.

4.a.2. Organically sequestered elements

Plankton are known to concentrate certain trace elements from sea water (Vinogradov, 1953; Goldberg, 1957; Martin & Knauer, 1973; and Eisler, 1981). Accordingly, dissolved concentrations of metals in the photic zone are very low. In the modern ocean, dissolved trace metal contents increase from a minimum value near the surface with depth as a result of oxidation of planktonic organic matter and the release of the metals back into the water column (Wilde & Berry, 1986). However, in anoxic basins such as the Black Sea, metal concentrations in the water column do not increase below the oxic layer (Brewer & Spencer, 1974). Thus, for an anoxic pycnocline and a shallow mixed layer predicted for warm climates in the Lower Paleozoic (Berry & Wilde, 1978; Wilde & Berry, 1982, 1984, Wright, Schrader & Holser, 1987) metal-rich organic matter may reach the bottom without extensive loss of its metal content. In particular, V, U and Mo in marine black shales show a high correlation with organic matter (Vine & Tourletot, 1970; Tardy, 1975; Brongersma-Saunders, 1966; Brumsack, 1986; and Degens et al. 1986). The combination of organic matter raining down from the high productivity zone along the outer shelf into transgressing anoxic waters could produce the conditions necessary for the transfer into bottom sediments of signature elements originally concentrated in organisms. Thus, the organic matter in the sediment, preserved in transit by anoxicity in the overlying water, would be a source of biologically concentrated trace metals. During deposition in anoxic bottom waters and a low rate of sedimentation, preserved organic material may sequester additional elements directly from the anoxic bottom waters. Anoxic processes such as sulfate reduction would produce sufficient anions so that mineral equilibrium reactions would permit transfer of the dissolved or sequestered metallic cations to sulfide and other anoxic mineral phases during diagenesis (Berner, 1981) and insure their preservation in shales.

4.a.3. Volatile and other elements

The source of the other enriched signature elements (As, Sb, Ba) is not as obvious (Brongersma-Saunders, 1966). Elements such as As and Sb are naturally volatile and occur with volcanic or hydrothermal activity. Holland (1979) suggested direct concentration of trace metals from an enriched sea water layer. Leventhal & Hosterman (1982) found it difficult to model the sequesting of metals directly from present sea water and suggested additional concentration by leaching of volcanic ash. Graphite with various antimonides have been found in sediment traps near hydrothermal vents on the modern East Pacific Rise (Jedwab & Boulegue, 1984). Gee (1980) and Bjørlykke (1974a) both have noted the presence of volcanic activity in the early Ordovician in Balto-Scandia.

4.a.4. Origin of the tripartite signature

The redox conditions and the high primary productivity may be linked. High productivity would deplete oxygen in the upper mixed layer rapidly so that the top of the anoxic zone would be shallower than in areas of lower productivity. Shoaling of the top of the anoxic layer also would occur during the vertical advection of upwelling. This would increase the area of the shelf intersected by the anoxic layer and further enhance the possibility of leaching of volcanic material. Thus, the sources of the anomalous high signature elements may be high surface productivity augmented by regional volcanism or hydrothermal activity venting directly into or leached by an oceanic anoxiclayer, where metals complexes would be preserved. Accordingly, the complete tripartite geochemical signature appears to be the result of a combination of special biological (high primary productivity), geological (marine volcanic or hydrothermal activity), and chemical oceanographic (particular anoxic zone) conditions relatively unique in the geologic record.

Due to the geologic and oceanographic rarity of the combination of conditions required to produced the unusual geochemical signature, we would assign New Brunswick to the Balto-Scandian block in the Tremadoc.

4.b. Non-signature deposition

4.b.1. Southern lapetus

Tremadoc areas with D .flabelliforme, but without the signature, such as Belgium and Wales, at approximately the same palaeolatitude as New Brunswick, may have been in (1) a regime of higher sedimentation (lower net organic matter in sediments) as compared with Balto-Scandia (Bjørlykke, 1971, 1974b;Anderson et al. 1979, pp. 83-4) which would dilute the organic matter and its related signature elements and (2) a different redox zone (more oxic or with more carbonate with higher Mn and Fe) preventing sequestering of the signature elements. Non-signature D. flabelliforme-bearing shales in Belgium and Wales suggests that the geochemical signature could be associated with outer shelf-upper slope deposits deeper and seaward of the Tremadoc position of the Belgium Wales locales. Thus, the New Brunswick locales with the signature could have been the deeper or seaward equivalents of the Maguma Terrane of Nova Scotia (Schenk, 1983), if the Tremadoc Nova Scotia, Wales, and Belgium were positioned as shown by Cocks & Fortey (1982). The lack of the volcanic and other signature elements in shales from Belgium and Wales compared with New Brunswick suggests a different oceanic chemistry. A geochemically and oceanographically simpler explanation would be that the complete signature would be geographically restricted to shales deposited on the western BaltoScandian block under relatively uniform source and oceanographic conditions.

However, a modified signature might be produced on the outer shelf with lower productivity, less volcanism, and different redox conditions. Potentially, Lower Ordovician equivalents with a modified signature might still be preserved along the Iapetus suture in Southern Newfoundland (Avalon), Central Ireland and the Lake District of England.

4.b.2. Northern lapetus

Figure 1 shows that the northern side of Iapetus would lie near or within the zone of equatorial Ekman planetary divergence. This also is region of high productivity by analogy with modern oceanic productivity zones (LaFond & LaFond, 1971). However, D. flabelliforme-bearing shales from New York, Levis (Berry et al. (1986) and Gaspe, Quebec do not show the organic Balto-Scandian signature (high V, Mo, U). Their geographic position during the Tremadoc suggests that these areas were not subjected to the enhanced upwelling caused by off shore advection which would have occurred off the west coast of Balto-Scandia. This may be due to their proximity to the main Laurentian tropical landmass which might generate (1) fresher water El Nino-like low productivity conditions in the surface mixed layer, and (2) parallel to the coast monsoonal currents impeding cross-shelf advection and upwelling. With no land vegetation in the Ordovician, runoff into the ocean from areas of tropical high precipitation could be vastly greater than seen today. This might produce low salinity conditions and reduce marine productivity even in areas of upwelling. However, northern Newfoundland, Ireland, and Scotland, as off shore islands, would be in the proper position as the equatorward flowing main current deflects westward along the equatorial divergence. The lack of preservation of signature or modified signature shales from Ireland and Scotland on the equatorial northern side of Iapetus may be explained by the subsequent destruction of the outer shelf margin by obduction during the Arenig and the emplacement of ophiolites along the north shore of Iapetus (Dewey, 1971; Church & Geyer, 1973; and Bluck et al. 1980). Future analyses of black shale samples from western Newfoundland (Fortey & Skevington, 1980), which was located in the Tremadoc equatorial belt would test our reconstruction and interpretations.

5. Post-Tremadoc black shale deposition in Iapetus

The absence of the signature in post Tremadoc Palaeozoic black shales deposited in the Iapetus Ocean, particularly in Balto-Scandia (Bjorlykke, 1974a,b; and Berry et al. 1986), may be the result of the closure of Iapetus and the northward movement of the landmasses on the south side of Iapetus away from the planetary divergence. This is suggested in post Early Tremadoc rocks of the Ceratopyge Shales near Oslo, which show a modification of the D. flabelliforme signature (Berry et al. 1986 p. 47). By Arenig time, Cocks & Fortey (1982) show BaltoScandia well north of the divergence. Thus, the outer shelf would not be as highly productive even if anoxic waters were on the shelf.

Acknowledgements. We thank the Sedgwick Museum, Cambridge University (European and Canadian); and K. Lindholm, University of Lund (Swedish) for providing samples for this study. Bolivia California Petroleum Company provided the Bolivian samples. In Norway, G. Henningsmoen and N. Spjeldnaes provided useful information on sampling locations. At the Los Alamos National Laboratory, L. R. Quintana assisted with the sample preparation and S. R. Garcia and K. H. Abel provided the elemental abundance using the automated neutron activation analyses system. M. Krup co-ordinated the production of the manuscript and did the illustration. The study was partially supported by the Institute of Geophysics and Planetary Physics, University of California as part of the Black Shale programme. This is contribution MSG-87-004 of the Marine Sciences Group.


ANDERSSON, A., DAHLMAN, B. & GEE, D. G. 1982. Kerogen and uranium resources in the Cambrian alum shales of the Billingen-Falbygden and Narke areas, Sweden. Geologiska Foreningens (i) Stockholm Fohandlingar 10, 197-209.

ANDERTON, R., BRIDGES, P. H., LEEDER, M. R. & SELLWOOD, B. W. 1979. A Dynamic Stratigraphy of the British Isles. London: George Allen and Unwin, pp. 83 4.

ARMANDS, G. 1972. Geochemical studies of Uranium, Molybdenium and Vanadium in a Swedish Alum shale. Stockholm University contribution in Geology 27, 1-148.

BERNER, R. A. 1981. A new geochemical classification of sedimentary environments. Journal of Sedimentary Petrology 51, 359-65.

BERNER, R. A. & RAISWELL, R. R. 1983. Burial of organic carbon and pyrite sulfur in sediments over Phanerozoic time: a new theory. Geochimica et Cosmochimica Acta 47, 855-62.

BERRY, W. B. N. & WILDE, P. 1978. Progressive Ventilation of the Oceans - an explanation for the distribution of the Lower Paleozoic black shales. American Journal of Science 278, 257-75.

BERRY, W. B. N., WILDE, P., QUINBY-HUNT, M. S. & ORTH, C. J. 1986. Trace Elements Signatures in Dictyonema Shales and their geochemical and stratigraphic significance. Norsk geologisk Tidsskrift 66, 45-51.

BERRY, W. B. N., WILDE, P.& QUINBY-HUNT, M. S. 1987 in press. The Oceanic Non-Sulfidic Oxygen Minimum Zone - A Habitat for Graptolites. Bulletin of the Geological Society of Denmark.

BJØRLYKKE, K. 1971. Petrology of Ordovician sediments from Wales. Norges geologisk Tidsskrift 51, 123-39.

BJØRLYKKE, K. 1974a. Depositional history and geochemical composition of Lower Paleozoic epicontinental sediments from the Oslo region. Norges geologiske Undersokelse 305, 1-81.

BJØRLYKKE, K. 1974b. Geochemical and mineralogical influence of Ordovician island arcs on epicontinental clastic sedimentation: a study of Lower Palaeozoic sedimentation in the Oslo region, Norway. Sedimentology 21, 251 72.

BLUCK, B. J., HALLIDAY, A. N., AFTALION, M. & MACINTYRE, R. M. 1980. Age and origin of Ballantrae Ophiolite and its significance to the Caledonian Orogeny and Ordovician time scale. Geology 8, 492-95.

BREWER, P. G. & SPENCER, D. W. 1974. Distribution of some trace metals in the Black Sea. In The Black Sea- Geology, Chemistry, and Biology (ed E. T. Degens and D. A. Ross) pp. 137-43. Tulsa: American Association of Petroleum Geologists Memoir 20.

BRONGERSMA-SANDERS, M. 1966. The fertility of the sea and its bearing on the origin of oil. In Advances in Organic Geochemistry (ed. G. D. Hobson and G. L. Speers) pp. 231-6. Oxford: Pergamon.

BRUMSACK, H. J. 1986. The inorganic geochemistry of Cretaceous black shales (DSDP Leg 41) in comparison to modern upwelling sediments form the Gulf of California. In North Atlantic Palaeocenaography (ed. C. J. Summerhayes and N. Schackleton). pp. 447-62, Oxford: Blackwells.

CHURCH, W. R. & GAYER, R. A. 1973. The Ballantrae ophiolite. Geological Magazine 110, 497-592.

COCKS, L. R. M. & FORTEY, R. A. 1982. Faunal evidence for oceanic separations in the Palaeozoic of Britain. Journal of the Geological Society of London 139, 465-78.

CONTE, M. H., BISHOP, J. K. B. & BACKUS, R. H. 1986. Nonmigratory 12-kHz, deep scattering layers of Sargasso Sea Origin in warm-core rings. Deep-Sea Research 33, 1869-84.

COOPER, R. A. 1979. Sequence and correlation of Tremadoc graptolite assemblages. Alcheringa 3 7-19.

DEGENS, E. T., EMEIS, K.-C., MYCKE, B. & WIESNER, M. G. 1986. Turbidites, the principal mechanism yielding black shales in the early deep Atlantic. In North Atlantic Palaeoceanography (ed C. J. Summerhayes and N. Shackleton). pp. 447-62, Oxford: Blackwells.

DEWEY, J. F. 1971. A model for the Lower Palaeozoic evolution of the southern margin of the early Caledonides of Scotland and Ireland. Scottish Journal of Geology 7, 219 40.

EISLER, R. 1981. trace Metal Concentrations in Marine Organisms. Oxford: Pergamon Press, 687 pp.

ERDTMANN, B.-D. 1982. Palaeobiogeography and environments of planktic dictyonemid graptolites during the earliest Ordovician. In The Cambrian-Ordovician boundary: sections, fossil distributions, and correlations eds. M. G. Bassett and W. T. Dean), pp. 9-27, Cardiff: National Museum of Wales, Geological Series No. 3.

ERDTMANN, B.-D. 1984. Outline eostratigraphic analysis of the Ordovician graptolite zones in Scandinavia in relation to the palaeogeographic disposition of the Iapetus. Geologica et Palaeontologica 18, 9-15.

FORTEY, R. A. 1984. Global earlier Ordovician transgressions and regression and their biological implications. In Aspects of the Ordovican System (ea. D. L. Brunton), pp. 37-50, Oslo: Palaeontological Contributions from the University of Oslo, No. 295.

FORTEY, R. A. & SKEVINGTON, D. 1980. Correlation of Cambrian-Ordovician boundary between Europe and North America: new data from western Newfoundland. Canadian Journal of Earth Sciences 17, 382-8.

FROELICH, P. N., KLINKHAMMER, G. P., BENDER, M. L., LUEDTKE, N. A. HEATH, G. R., CULLEN, D., DAUPHIN, P., HAMMOND, D., HARTMAN, B. & MAYNARD, V. 1979. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochimica et Cosmochimica Acta 43, 1075-90.

GEE, D. G. 1980. Basement-cover relationships in the central Scandinavian Caledonides. Geologiska Foreningens (i) Stockholm Forhandlingar 102, 455 74.

GEE, D. G. 1981. The Dictyonema-bearing phyllites at Nordaunevoll, eastern Trondelag, Norway. Norsk geologisk Tidsskrift 61, 93-5.

GOLDBERG, E. D. 1957. Biochemistry of trace metals. In Paleoecology (ea. J. W. Hedgpeth), pp. 345-57, New York: Geological Society of America Memoir 67, 1.

HOLLAND, H. D. 1979. Metals in black shales - a reassessment. Economic Geology 74, 1676-80.

JEDWAB, J. & BOULEGUE, J. 1984. Graphite crystals in hydrothermal vents. Nature 310, 41-3.

LANDING, L., TAYLOR, M. E. & ERDTMANN, B.-D. 1978. Correlation of the Cambrian-Ordovician boundary between the Acado-Baltic and North American faunal provinces. Geology 6, 75-8.

LA FOND, E. C. & LA FOND, K. G. 1971. Oceanography and its relation to marine production. In Fertility of the Sea, (ea. J. D. Costhlow), pp. 241-65. New York: Gordon and Breach.

LEGRAND, PH. 1974. Development of rhabdosomes with four primary branches in the group Dictyonema flabelliforme (Eichwald). In Graptolite studies in honour of O. M. B. Bulman. (eds. R. B. Rickards, D. E. Jackson and C. P. Hughes), pp. 19-34, London: The Palaeontological Association Special Papers in Palaeontology No. 13.

LEVENTHAL, J. S. & HOSTERMAN, J. W. 1982. Chemical and Mineralogical analysis of Devonian Black-shale samples from Martin County, Kentucky; Caroll and Washington Counties, Ohio; Wise County, Virginia; and Overton County, Tennessee, USA. Chemical Geology 37, 239 64.

LINDSTROM, M. & VORTISCH, W. 1983. Indications of upwelling in the Lower Ordovician of Scandanavia. In Coastal Upwelling (eds. J. Thiede and E. Suess) Part B., pp. 535-51, New York: Plenum.

McGOWAN, J. A. 1972. The Nature of Oceanic Ecosystems. In The Biology of the Oceanic pacific (ea. C. B. Miller), pp. 9-28, Corvallis, Oregon: Oregon State University Press.

MARTIN, J. H. & KNAUER, G. A. 1973. The elemental composition of plankton. Geochimica et Cosmochimica Acta 37, 1639-53.

MINOR, M. M., HENSELY, W. K., DENTON, M. M. & GARCIA, S. R. 1982. An automated activation analysis system. Journal of Radioanalytical Chemistry 70, 459-771.

PARRISH, J. T. 1982. Upwelling and Petroleum Source Beds, with Reference to Paleozoic. American Association of Petroleum Geologists Bulletin 66, 750 74.

PELTOLA, E. 1968. Geochemical features in black schists of the Outokumper area, Finland. Bulletin of the Geological Society of Finland 40, 39-50.

RATSWELL, R. & BERNER, R. A. 1986. Pyrite and organic matter in Phanerozoic normal marine shales. Geochimica et Cosmochimica Acta 50, 1967-76.

RAISWELL, R. & BERNER, R. A. 1987. Organic carbon losses during burial and thermal maturation of normal marine shales. Geology 15, 853-6.

REID, J. L. JR. 1985. Intermediate Waters of the Pacific Ocean. Baltimore: John Hopkins University Press, 85 pp.

SCHENK, P. E. 1983. The Meguma Terrane of Nova Scotia, Canada - an aid in Trans-Atlantic correlation. In Regional Trends in the Geology of the AppalachianCaledonian-Hercynian-Mauritanide Orogen (ed. P. E. Schenk), pp. 121-30, Dordrecht, Holland: D. Reidel.

SCOTESE, C. R., BAMBACH, R. K., BARTON, C., VAN DER VOO, R. & ZEGLER, A. M. 1979. Paleozoic basemaps. Journal of Geology 87, 21-77.

SCOTESE, C. R. 1986. Phanerozoic Reconstructions: A New Look at the Assembly of Asia. University of Texas Institute for Geophysics Technical Report No. 66, 54 pp.

STOMMEL, H. 1957. A survey of ocean current theory. DeepSea Research, 4, 149-84.

STUMM, W. & MORGAN, J. J. 1970. Aquatic Chemistry New York: Wiley, 583 pp.

SUNBLAD, K. & GEE, D. G. 1985. Occurrence of an uraniferous-vanadiniferous graphitic phyllite in the Koli Nappes of the Stekenjokk area, central Swedish Caledonides Geologiska Foreningens (i) Stockholm Fohandlingar 106, 269-74.

TARDY, Y. 1975. Element partition ratios in some sedimentary environments, I Statistical Treatments, II Studies on North-American black Shales. Sciences Geologiques: Bulletin, Srasbourg 28, 75-95.

VINE J. D. & TOURTELOT, E. B. 1970. Geochemistry of black shale deposits - a summary report, Economic Geology 65, 253-72.

VINOGRADOV, A. P. 1953. The elemental composition of marine organisms. New Haven: Sears Foundation for Marine Research Memoir 2, 647 pp.

WEDEPOHL, K. H. 1964. Untersuchungen am Kupferschiefer in Nordwestdeutschland; Ein Beitrag zur Deutung der Genese bituminOser Sedimente. Geochimica et Cosmochimica Acta 28, 305-64.

WHITTINGTON, H. B., DEAN, W.T., FORTEY, R. A., RICKARDS, R. B. RUSHTON, A. W. A. & WRIGHT, A. D. 1984. Definition of the Tremadoc Series and the series of the Ordovician System in Britain. Geological Magazine 121, 17-33.

WIEBE, P. H. & McDOUGALL, T. J. 1986. Introduction to a collection of papers on warm-core rings. Deep-Sea Research 33, 1455-7.

WILDE, P. 1987. Model of Progressive Ventilation of the Late Precambrian-Early Paleozoic Ocean. American Journal of Science 287, 442-59.

WILDE, P. & BERRY, W. B. N. 1982. Progressive Ventilation of the Oceans II, potential for return to anoxic conditions in the post-Paleozoic. In Nature and Origin of Cretaceous Carbon-Rich Facies, (eds. S. O. Schlanger and M. B. Cita), pp. 209-24, New York: Academic Press.

WILDE, P. & BERRY, W. B. N. 1984. Destabilization of the Oceanic density structure and its significance to Marine " Extinction " events. Palaeogeography, Palaeoclimatology, Palaeoecology 48, 143-62.

WILDE, P. & BERRY, W. B. N. 1986. Role of Oceanographic factors in the generation of Global Bio-Events. In Global Bio-Events, (ea. O. Walliser), pp. 75-91, Berlin: Springer-Verlag.

WILDE, P., QulNsY-HuNT, M., BERRY, W. B. N. & ORTH, C. J. 1984. Anoxic facies in the Lower Paleozoic Ocean. Geological Society of America Annual Meeting Reno pp. 994.

WILDE, P., BERRY, W. B. N., QUINBY-HUNT, M. S., ORTH, C. J., QUINTANA, L. R. & GILMORE, J. S. 1986. Iridium abundances across the Ordovician-Silurian Stratotype. Science 233, 339 41.

WOOSTER, W. S. & REID, J. L., JR. 1963. Eastern Boundary Currents. In The Seas ll, Composition of Sea Water and Descriptive Oceanography, (ed. N. H. Hill), pp. 253-80. New York: Wiley Interscience.

WRIGHT, J. SCHRADER, H. & HOLSER, W.T. 1987. Paleoredox variations in ancient oceans recorded by the rare earth elements in fossil apatite. Geochimica et Cosmochimica Acta 51, 631-44.

YAMAMATO, T. & NISHIZAWA, S. 1986 Small-scale zooplankton aggregationat the front of a Kuroshiro warm-core ring. Deep-Sea Research 33, 1729-40.