THE PROVENANCE OF LOW-CALCIC BLACK SHALES
M.S. QUINBY-HUNT
 P. WILDE
 and 
W.B.N. BERRY
Marine Sciences Group 
University of California
                Berkeley, CA 94720, USA.
After Mineralum Deposita, vol. 26, p. 113-121 (1991)
Modified and revised for the World Wide Web:  April, 1997

The elemental concentration of sedimentary rocks depends on the varying reactivity of each element as it goes from the source through weathering, deposition, diagenesis, lithification, and even low rank metamorphism. However, non-reactive components of detrital particles ideally are characteristic of the original igneous source and thus are useful in provenance studies. To determine the source of detrital granitic and volcanic components of low- calcic (< 1 % CaCO₃) marine black shales, the concentrations of apparently non-reactive (i.e. unaffected by diagenetic, redox and/or low-rank metamorphic processes) trace elements were examined using standard trace element discrimination diagrams developed for igneous rocks. The chemical data was obtained by neutron activation analyses of about 200 stratigraphically well-documented black shale samples from the Cambrian through the Jurassic. A La-Th-Sc ternary diagram distinguishes among contributions from the upper and bulk continental crust and the oceanic crust (Taylor and McLennan, 1985). All the low-calcic black shales cluster within the region of the upper crust. Th-Hf-Co ternary diagrams also are commonly used to distinguish among the upper and bulk continental crust and the oceanic crust (Taylor and McLennan, 1985). As Co is redox sensitive in black shale environments, it was necessary to substitute an immobile element (i.e. example Rb) in the diagram. With this substitution the black shales all cluster in the region of the upper continental crust. To determine the provenance of the granitic component (Pearce et al., 1984), plots of Ta vs Yb and Rb vs Yb+Ta shows a cluster at the junction of the boundaries separating the volcanic arc granite (VAG), syn-collision granite (syn-COLG), and within-plate granite (WPG) fields. The majority fall within the VAG field. There are no occurrences of ocean ridge granite (ORG). The minimal contribution of basalts to marine black shales is confirmed by the ternary Wood diagram Th- Hf/3-Ta (Wood et al., 1979). The black shales plot in a cluster in a high Th region outside the various basalt fields, which suggests contribution from the continental crust.

INTRODUCTION

Black shales comprise a widely-found, unique rock suite, particularly in the pre-Jurassic. Many black shales are hosts for metals (Swanson, 1961; Vine and Tourtelot, 1970; Tourtelot, 1970; Armands, 1972; Bjørlykke, 1974; Holland, 1979; Gee, 1980; Leventhal and Hosterman, 1982; Andersson et al., 1982; Grauch and Leventhal, 1989). Others are source rocks for petroleum (Brooks and Fleet, 1987). Nearly all contain distinctive fossils (Bulman, 1955; Berry et al., 1989), lack signs of bioturbation and are characteristically thinly laminated. Most black shales accumulated in oxygen-poor to anoxic environments (Pettijohn, 1949; Berry and Wilde, 1978). Most time- synchronous, geographically wide spread marine Paleozoic-Mesozoic black shales formed on the outer parts of the continental shelf or on the upper continental slope (Berry and Wilde, 1978; Hallam, 1987). Modern open ocean environments that may serve as analogues for ancient marine shelf and slope black shale environments of deposition include that off Peru-Chile and the southern California shelf basins (Berry et al., 1989; Schwalbach and Gorsline, 1985; Thornton, 1984). Sediments accumulating in these modern depositional environments include a rain of pelagic clays and organically-derived materials as well as clastic debris derived from nearby lands and carried down-slope into the basins by slumps and turbidity flows (Schwalbach and Gorsline, 1985; Degens et al., 1986). Because of both the economic and scientific interest (IGC Project 254: Metalliferous Black Shales, Grauch and Leventhal, 1989) in black shales, their sedimentary history, including their source or provenance, is of concern.

Although sources of modern sediments accumulating on shelves may be determined through observation using sediment traps, provenance of ancient (fossil) shelf sea sediments must be determined from a number of indirect lines of evidence. Geochemical elemental and mineralogic analysis combined with palaeogeographic reconstructions is proposed as a useful approach to determine the provenance of ancient black shale deposits. Because plate motions and related tectonic activity may distort or destroy much of the original depositional environments, provenance of the remnants of these environments may have to be deduced from the chemical evidence left in the rock record.

To test this hypothesis, we have used 200 stratigraphically well-documented black shales (Quinby-Hunt et al., 1989) from the Paleozoic and Mesozoic which were analyzed for 40 elements using neutron activation (see Appendix A for analytical discussion). C/S ratios have been used as environmental indicators (Berner and Raiswell, 1983; Raiswell and Berner, 1986). However, C and S are highly mobile elements and would be entirely unsuitable for determining the provenance the detrital component of black shales. In fact, techniques for determining C and S require unweathered samples, such as core material, to yield reliable results due to the mobility of C and S (Raiswell and Berner, 1986). Such optimal conditions are difficult to find in the field and most palaeontologically-dated material from well-documented collections made from outcrops are not acceptable for C/S analyses. The major fossiliferous black shale sedimentary intervals occur in the Paleozoic and Mesozoic. Thus the carbon content of these rocks may be more indicative of age or maturity than that of depositional environmental conditions (Raiswell and Berner, 1986).

PREVIOUS USE OF CHEMICAL VARIATION DIAGRAMS

Taylor and McLennan (1985) determined provenance of certain post-Archean shales in Australia by using ternary diagrams involving two systems: La-Th-Sc and Th-Hf-Co. Igneous petrologists also have used chemical variation of trace metals to ascertain the tectonic relationships of various magmas. Detrital fractions as erosion products from such igneous rocks potentially would carry the same signature and could be used in provenance studies if the indicated trace metals survived erosion and deposition. Wood and others (1979) used concentrations of Ta, Hf, La and Th to distinguish between mid-ocean ridge basalts and those that erupted at active plate margins. Winchester and Floyd (1977) differentiated between different magma series using concentrations of Ti, Zr, Y, Nb, Ge, Ga, and Sc as indicators of melt composition. Pearce and others (1984) used Ta-Yb and Rb-Ta+Yb diagrams to discriminate between syn-collision, within-plate, volcanic arc, and ocean ridge sources. Inasmuch as most of the black shale samples analyzed in this study came from sites that were on active plate margins or from continental slope-fan environments close to volcanic activity, the geochemical data obtained in this study were plotted on the diagrams from these earlier studies to address the provenance of the samples.

When examining the provenance of sedimentary rocks, it is essential to use for discrimination only those elements whose concentrations are unaffected by the physical and chemical conditions prevalent during erosion, redeposition, diagenesis, and to a degree low-rank metamorphism (Winchester and Floyd, 1977; Wood et al., 1979). The samples used in this study are true shales and not slates. Thus, at a maximum, they have been subjected only to low-rank metamorphism. Accordingly, the elemental content would not have been modified as described by Haack et al. (1984) for pelites during regional metamorphism.

Conventional elemental analyses by wet chemical methods for elements such as Al, Si, Ca etc. only yielded major elements whose sources were too ubiquitous or were not immobile in the diagenetic process to be much use in sedimentary provenance studies. However, the development of high precision analytical techniques run automatically with large batches of samples provided an additional data base of trace elements which may be used in more refined provenance study. Elements commonly considered immobile, that is insensitive to secondary processes, include Zr, Nb, Ti, Sc, Y, P, Th, Ta, Hf, and the REEs (Winchester and Floyd, 1977; Wood et al., 1979; Taylor and McLennan, 1985). Studies of the modern distribution of REE's in sea water and sediments suggest the potential for diagenetic mobility (Piper, 1974; Tlig and Steinberg, 1982; Sholkovitz, 1989; German and Elderfield, 1990; Stanley and Byrne, 1990), particularly in organic-rich sediments. Baker and deGroot (1983) indicate high REE mobility for felsic volcanics in ancient (1.8 Ga) sea water. Also Michel (1984) showing modification of U and Th during in place weathering of granites. The extent of mobility of these elements in relation to black shales can be studied by the use of the discrimination diagrams.

BLACK SHALE SAMPLES

The samples used in this study are primarily from Ordovician and Silurian stratal units. Included as well are some samples of Cambrian, Carboniferous and Jurassic age (Quinby-Hunt et al., 1989, p.9). This study is strongly biased towards Cambrian-Silurian samples (176 of 206) from sites that were on the margin of the Iapetus Ocean (Scotese and McKerrow 1990), which include parts of modern Canada, eastern United States, Great Britain, Balto-Scandia, and Western Europe. All samples collected for this study were identified, stratigraphically and palaeontologically, by W.B.N. Berry or were in well-documented collections (for example, Sedgwick Museum, Cambridge; Museum of Paleontology, Berkeley). The study focused on field-collected black shales, and were limited to samples that contain less than 1% CaCO₃ and greater than 50000 ppm Al. This study specifically excluded shaley limestones or limy shales (sensu Pettijohn, 1949, p.291), in order to focus on the composition of the argillaceous black shale facies.

In order to demonstrate the concentration variability, Figure 1 shows, for each element important to the discrimination schema used in this study, the frequency distributions, as percent, plotted versus the midpoints of distribution bins used in determining the distribution. Each bin has boundaries of 1.2, 1.8, 2.6, 3.8, 5,6, 8.3, 12 and so on. Boundaries were chosen for consistency with the geochemical analyses of North American black shales by Vine and Tourtelot (1970). A unimodal frequency distribution with a narrow range of concentrations suggests a well-defined depositional environment with respect to that element. For this set of samples, the Ti (Figure 1) and Al distributions are unimodal. These elements are associated with detrital sources (clays and feldspars), so their unimodal distribution suggests that the sample set was indicative of a shale facies as opposed to a chemical (carbonate or evaporate) facies. Other unimodal elements were Sc, Ga, Rb, Zr, Hf, Ta, Th, and most of the rare earth elements, including La, Ce and Yb (REEs). In contrast, Co has a complex frequency distribution (Figure 1) indicating that within the black shale depositional environment, this element varies independently of the basic detrital shale component. Other elements with pronounced multimodality included V, Mn, Fe, As, Se, Mo, Sb, Ba, and U (Quinby-Hunt et al., 1988; Quinby-Hunt et al. 1989). Variation in the concentration of these elements may be associated with redox conditions, biological activity or presence of organic matter, or volatile/volcanic oceanic sources.

DISCUSSION

Multi-modal frequency distributions show that the black shale facies is chemically complex and may contain several chemo-facies. Ideally, if the black shale facies observed in the field was a single chemical-sedimentary quantity, the elements characteristic of that facies would show a unimodal distribution. When compared with the classic shale composites (Turekian and Wedepohl, 1961; Gromet et al., 1984; Taylor and McLennan, 1985) and with the study of black shales by Vine and Tourtelot (1970), this study finds excellent agreement for the elements generally associated with detrital minerals, such as Al and Ti. The absolute concentration of the various clay minerals within the shales may be diluted to varying degrees by the presence of other minerals, such as silicates. The general provenance of black shales may be investigated by comparison with the composition of various source rocks by using techniques developed to investigate the provenance of those sources.

La-Th-Sc, Th-Hf-Co Ternary Diagrams - Discrimination of Shales.
Taylor and McLennan (1985) used La-Th-Sc and Th-Hf-Co ternary diagrams to investigate provenance of several Post-Archean Australian shales (PAAS). This approach assumes that the composition of these trace elements may vary only by mixing with unmodified detrital minerals reflective of the composition of the various igneous source rocks. Taylor and McLennan (1985) use as end-members the upper continental crust and oceanic crust. The La-Th-Sc diagram (Figure 2) illustrates the similarity of the composition of detrital elements in shales to that of the upper continental crust. The field formed by the more than 200 samples making up our low-calcic black shale data base superimposes with that of the PAAS (Figure 2) for La-Th-Sc and generally falls between Taylor and McLennan's (1985) approximation of the upper crust and the bulk continental crust. Therefore, indicator detrital elements of the black shales and the post-Archean shales show the expected upper "continental" source, using the Taylor and McLennan (1985) model of andesitic (continental) and basaltic (oceanic) crust as mixing end members.

The Th-Hf-Co ternary diagram (Figure 3), used by Taylor and McLennan (1985) for their PAAS, shows that our black shales differ significantly from the Post-Archean shales, and illustrates the care one must take when assuming an element is immobile under erosional, depositional and digenetic environments. The black shale field, of Cambrian through Jurassic age, extends far beyond that of PAAS. The field also extends beyond the source rocks. It is unlikely this difference is a result of the black shales studied herein and "Post-Archean" shales having a different detrital provenance as both are Post-Archean. Certainly no difference is indicated by the La- Th-Sc plot. Thus one or more of the trace-metal end-members of Figure 3 must undergo post-erosional chemical changes and thus are not immobile under conditions of black shale deposition and diagenesis. Co has a wider range of concentration in the black shales than in the PAAS. Co does not behave as an immobile element in varying redox conditions; its concentration varies with that of Fe and Mn under the differing conditions observed in the Dictyonema shales of Balto-Scandia (Berry et al. 1986; Wilde et al., 1989). Under the chemical (reducing or biological) conditions which could occur either in the water column or in early diagenesis (Quinby-Hunt et al., 1988), Co may be mobilized. To test this hypothesis, Rb is substituted for Co in a Th-Hf-Rb/10 ternary diagram (Figure 4). Rb is generally considered to be immobile and is used by both Wood (Wood et al., 1979; Wood, 1980) and Pearce et al. (1984) in their discrimination schema. In Figure 4, both the black shale composite and the PAAS plot in an overlapping field. The wide variation of Co in black shales compared with the small variation of Co in the PAAS shown in Figure 3 suggests that the black shale facies (sensu latto) encompasses a wider variation in redox conditions than seen the samples of the PAAS or other general shale composites. Conversely, the reason Co forms a cluster in a Th-Hf-Co "immobile" element ternary diagram for PAAS is that those shales have experienced relatively uniform chemical conditions, presumably oxic as the NASC an oxic composite plots well within the PAAS field on the Th-Hf-Co diagram. Under these conditions the Co concentration varies within small limits during erosion and deposition.

Zr/TiO₂ vs Ce or Ga - Discrimination of Magma Series.
Winchester and Floyd (1977) differentiate between primary magma series using the concentrations of the x-ray fluorescence-determined elements, Ti, Zr, Y, Nb, Ge, Ga, and Sc, as invariant indicators of initial igneous composition. They found that a plot of Zr/TiO₂ ratio vs Ce can be used to discriminate between various volcanic igneous rock types as a function of their alkali metal content. Unfortunately, the neutron activation analyses were unable to detect Zr or Ga in most of the black shale samples. In addition, Ce, although readily detected using NAA, like Co, is mobile in marine environments (for example, Piper, 1974; German and Elderfield, 1990), thus we cannot determine the validity of using these discrimination diagrams from NAA data. Potentially, the Zr/TiO₂ ratio vs Ga might be useful with data obtained using other chemical techniques.

Ta-Yb and Rb-(Ta+Yb) Diagrams - Discrimination among Granitic Sources.
Pearce et al. (1984) use the NAA-determined elements, Yb, Ta, and Rb, to discriminate among what they termed "granitic" sources: ocean ridge "granites", volcanic arc "granites", within-plate "granites" and syn-collision "granites". They define ocean ridge granites (ORG) as those recovered from deep ocean dredging or from ophiolite complexes. Volcanic arc granites (VAG) encompass a wide variety of compositions ranging from the tholeiitic through the calc-alkaline to shoshonitic. They may be associated with ocean island arcs or with active continental margins, but include only those involving subduction of ocean crust. Within plate granites (WPG) intruded into the continental crust or oceanic crust. Collision granites are formed during (syn-COLG) continent-continent, continent-arc, or arc-arc collision. As Figures 5 and 6 show, the vast majority of the low-calcic black shales fall well within the volcanic arc granite field zone. A few samples appear in the border area for within plate granites or syn-collision granites, however, very few of these borderline samples appear in the WPG zone on both figures 5 and 6. No samples appear as syn-COLG in both figures 5 and 6. These diagrams suggest that the primary source for the materials of the black shales studied is from igneous rocks associated with volcanic arcs.

Hf vs Ta, La vs Ta, and Hf-Th-Ta Diagrams - Discrimination among Basalts.
Wood et al. (1979) have used the concentrations of Hf, Ta, La, and Th to discriminate between various mid-ocean ridge basalts, as well as distinguishing between basalts erupted at destructive plate margins, and those contaminated with continental crust. The Taylor and McLennan (1985) style diagrams above strongly suggest that an oceanic crustal source is not likely for the black shales. The Wood diagram permits discrimination among magmas formed at destructive margins, within plate magmas or those with continental crustal contamination.

As is expected from the Taylor and McLennan (1985) style diagrams above, the black shales studied show no affinity to the patterns demonstrated by mid-ocean ridge basalts (MORBs). Normal ridge basalts (N-type MORBs) are distinguished by La/Ta ca. » 15 and Hf/Ta ca. » 7. Oceanic ridges associated with more enriched basalts are dominated by tholeiitic basalts (E-type MORBs) and are identified as having Hf/Ta < 7, but >2; and La/Ta is about 10. The low-calcic black shales of this study clearly can not be associated with N-type or E-type MORBs with average La/Ta ratios > 50, and Th/Hf ratios < 1.25. The three samples whose La/Ta ratio suggested that they might have an ocean ridge source, are probably anomalous as none of the other discrimination diagrams indicate such as source for these samples. The black shales fall more reasonably within the categories associated with mid-plate basalts or basalts from destructive margins as their Hf/Ta ratios are < 2.

Discrimination of within-plate basalts from those erupted at destructive plate margins can be achieved using a Hf/3-Th-Ta ternary diagram (Wood et al., 1979; Thompson et al. 1980; Wood, 1980) (Figure 7). N-type and E-type MORBs tend to occur nearer to the Hf/3 apex or center (their fields A and B). Within plate basalts are depleted in Hf, but tend to fall equidistant between Th and Ta (Field C). Basalts erupted at plate margins and differentiates fall within field D due to crystal fractionation which pushes residual liquid into the Th apex of the triangle. Field D has been divided into a Hf-enriched segment associated with island arc vulcanism and a Hf- depleted section representative of calc-alkaline lavas. Average amphibolite falls in the Hf-depleted Th apex of the ternary diagram (Figure 7). The black shales fall near the Th apex of the D-field in Figure 7. The entire field of black shales is tightly clustered and falls well away from both the oceanic basalts and the within-plate basalts. Thus any basaltic contribution to the black shales composition appears to be derived from volcanic activity associated with active, destructive plate margins.

CONCLUSIONS

The low-calcic shales analyzed here show smaller variability of composition than was observed by Vine and Tourtelot (1970) for the unimodal elements because, in this study, the samples are basically clays and other detrital minerals with little carbonate to complicate the mineralogy. The multimodality seen in the low-calcic shales suggests that factors other than source rock composition operate during deposition and diagenesis effect the elemental composition of black shales. The use of discrimination diagrams leads to the conclusion that the majority of the low-calcic black shales in this sample set contained clays that accumulated on shelves of active or destructive plate margins. Plotting the data from the 200 low-calcic black shales on the Taylor-McLennan (1985) ternary diagrams indicates an upper continental crustal source. The data when plotted as suggested by Pearce et al. (1984) show derivation of clays formed in volcanic arc systems in which subduction of ocean crust is involved. A small number of samples may result from erosion from materials intruded into the oceanic or continental crust. If the data are plotted on the Wood (1980) ternary diagram, all samples cluster tightly within the region of calc-alkaline materials formed at destructive plate margins. The analyses and interpretations of the geochemical data from the ancient black shales are consistent in suggesting that the majority of the shales accumulated in an active plate margin tectonic setting.

When the positions of the shales are placed on palaeogeographic maps on which plate positions are indicated for intervals within the Paleozoic and Mesozoic periods (Scotese and McKerrow, 1990; Dewey, 1982), the majority of the samples fall on plate margins. Volcanic arc sources appear to be likely for the Balto-Scanian Dictyonema shales and for the Late Ordovician-Early Silurian shales from southern Scotland (Wilde et al, 1986; 1989). The Jurassic samples are from areas of subsiding shelf seas of the time (Scotese and McKerrow, 1990). The source for these shale samples presumably was continental material. Thus, the geochemical approach to determining the provenance of these shales is consistent with their likely palaeogeographic positions. The geochemical analyses may provide more precise information on provenance for the Paleozoic samples because the data on which those palaeogeographies are based are meager.

The clustering of data in defined tectonic fields in the diagrams depicted suggests, provenance of ancient shales may appropriately be determined from geochemical analyses, with proper concern for identification of immobile elements. Such analyses can suggest the dominant lithology and tectonic setting in the original drainage area for the samples, and, thereby, enhancing palaeogeographic reconstructions.

APPENDIX A

ANALYTICAL PROCEDURES

Neutron Activation Analysis. A suite of about 40 elements were determined (Quinby-Hunt et al., 1989, pp.10- 11). All elemental abundances were determined at a single laboratory, Los Alamos National Laboratory, using automated neutron activation analysis (Minor et al. 1982), minimizing the possibility for introduction of error commonly introduced by variations in analytical techniques or laboratory (Fairbairn et al., 1951; Quinby-Hunt et al., 1986). Intercalibration studies of several neutron activation facilities such as the Lawrence Berkeley Laboratory (F. Asaro, pers. comm.) have shown excellent agreement. Most of the elemental concentrations were determined using conventional reduction of gamma-ray spectra of the radioactive isotopes. Uranium was determined by delayed neutron counting. The automated system was calibrated with a collection of U.S. Geological Survey, National Bureau of Standards, and Canadian Geological rock standards. The system is checked periodically for stability against these standards. NAA provides isotopic concentrations which are translated to elemental concentrations. NAA can not be used to determine first row elements of the periodic table (Li through F) or S, Si, or P. Detection limits for Ni, Cu, Ag, Hg, were too high for determination in the black shales. Pb, Tl and Bi cannot be determined using NAA.

Detection Limits. For most elements, NAA involves the deconvolution of complex gamma-ray spectra. Minimum detection limits therefore are not absolute values, but depend strongly on the concentrations of the other constituents in the sample. For example, the levels of Mn encountered in black shales can interfere with the spectra of some elements counted after the first irradiation, such as Cl or Ca. Corrections were made for elements that are also fission products of U, for example, Ba, Mo, Zr, La, Ce, Nd and Sm. Elements for which a significant number of samples contained undetectable concentrations include: Cl, Ca, Cu, Zn, Ga, Se, Br, Sr, Zr, Mo and Nd. Of these elements, Ga and Zr, elements easily determined using x-ray fluorescence, are important to some discrimination schemes. Because for Zr and Ga, less than 50% of the samples did not contain detectable concentrations levels of an element, we calculated the mean of the detectable concentrations, but also note the percentage of all samples containing less than some appropriate concentration. Forty-seven percent of the samples contained measurable amounts of Ga. The average Ga concentration for samples in which Ca was detected was 22 ppm. However, for the rest of the samples, all that can be determined is that 81% contained less than 38 ppm. Only 35% of the samples contained detectable Zr. The average concentration was 230 ppm and 89% contain less than 380 ppm.

ACKNOWLEDGMENTS

The authors thank the numerous people and agencies who have provided or collection localities for this study, especially: B.-D. Erdtmann, K. Lindholm, G. Hemmingsmoen, N. Spjeldnaes, the Bolivia-California Petroleum Company, R.A. Raiswell, The Sedgwick Museum, Cambridge University. Carl Orth, S. Garcia, J.S. Gilmore, and L.R. Quintana of the Los Alamos National Laboratory performed the Neutron Activation Analysis. W. Huff alerted us to the usefulness of igneous discrimination diagrams through his and his colleagues' work with ash beds. We thank two anonymous reviewers whose careful reviews and comments were appreciated. M. Krup painstakingly prepared the illustrations. Support for this work was in part from the Institute of Geology and Geophysics, University of California as part of the Joint Black Shale Project between Berkeley and Los Alamos. This is contributions MSG-90-001 of the Marine Sciences Group, University of California, Berkeley.

REFERENCES

Andersson A., Dahlman, B., and Gee, D.G.: Kerogen and uranium resources in the Cambrian alum shales of the Billingen-Falbygden and Nørske areas, Sweden. Geol. Foren. Stockh. Forh. 102, 197-209 (1982).

Armands, G.: Geochemical studies of uranium, molybdenum and vanadium in a Swedish alum shale. Stockholm Univ. Contrib. Geol. 21, 1-148 (1972).

Baker, J.H. and de Groot, P.A.: Proterozoic seawater - Felsic volcanics interaction W. Berglagen, Sweden. Evidence for high REE mobility and implications for 1.8 Ga seawater compositions. Contrib. Mineral. Petrol. 82, 119-130 (1983).

Berry, W.B.N. and Wilde, P.: Progressive ventilation of the oceans - an explanation for the distribution of the Lower Paleozoic black shales. Amer. J. Sci. 278, 257-275 (1978).

Berry, W.B.N., Wilde, P., Quinby-Hunt, M.S. and Orth, C.J.: Trace element signatures in Dictyonema Shales and their geochemical and stratigraphic significance. Norsk Geologisk Tiddskrift 66, 45-51 (1986).

Berry, W.B.N., Wilde, P. and Quinby-Hunt, M.S.: Paleozoic (Cambrian through Devonian) anoxitropic biotopes. Palaeogeography, Palaeoclimatology, Palaeoecology 74, 3-13 (1989).

Berner, R.A. and Raiswell, R.: Burial of organic carbon and pyrite sulfur in sediments over Phanerozoic time: A new theory. Geochim. Cosmochim. Acta 47, 855-862 (1983).

Bjørlykke, K.: Depositional history and geochemical composition of Lower Paleozoic epicontinental sediments from the Oslo Region. Norske Geologisk Undersøkelse 305, 1-81 (1974).

Brooks, J. and Fleet, J. (Eds.) Marine Petroleum Source Rocks. Geological Society of London Special Publication 26, Oxford: Blackwell Scientific 1987.

Bulman, O.M.B.: Graptolithnia. In: Treatise on Invertebrate Paleontology, R.C. Moore, Ed., pp. V3- V109. Lawrence, Kansas: Geological Society and University of Kansas Press 1955.

Degens, E.T., Emeis, K.-C., Mycke, B. and Wiesner, M.G.: Turbidites, the principal mechanism yielding black shales in the early deep Atlantic Ocean. In: North Atlantic Palaeoceanography, C.P. Summerhayes and N.J. Shackleton, Eds., Geological Special Publication No. 21, pp. 361-376. Oxford: Blackwell Scientific 1986.

Dewey, J.F.: Plate Tectonics and the evolution of the British Isles, J.Geol.Soc. London 139, 371-412 (1982).

Fairbairn, H.W. and others: A cooperative investigation of precision and accuracy in chemical, spectrochemical and modal analysis of silicate rocks. In: Contributions to Geochemistry, 1950-51, pp.1- 71. Geological Society Bulletin 980. Washington, DC: U.S. Government Printing Office 1951.

Gee, D.G.: The Dictyonema-bearing phyllites at Nordaunevoll, eastern Trondelag, Norway. Norske Geologisk Tiddskrift 61, 93-95 (1980).

German, C.R. and Elderfield, H.: Rare earth elements in the NW Indian Ocean. Geochim. Cosmochim. Acta 54, 1929-1940 (1990).

Grauch, R.I. and Leventhal, J.S. (Eds.): Metalliferous Black Shales and Related Ore Deposits, U.S.Geological Survey Circular 1037. Washington, DC: US Government Printing Office, 1989.

Gromet, L.P., Dymek, R.F., Haskin, L.A., and Korotev, R.L.: The "North American shale composite": Its compilation, major and trace element characteristics. Geochim. Cosmochim. Acta 48, 2469-2482 (1984).

Haack, U., Heinrichs, H., Boness, M., and Schneider, A. Loss of metals from pelites during regional metamorphism. Contrib. Mineral. Petrol. 85, 103-115 (1984).

Hallam, A.: Mesozoic marine organic-rich shales. In: Marine Petroleum Source Books, J. Brooks and A.J. Fleet, Eds. pp. 251-261. Geological Society Spec. Pub. 26. Oxford: Blackwell Scientific 1986.

Holland, H.D.: Metals in black shales - A reassessment. Econ. Geol. 74, 1676-1680 (1979)

Leventhal, J.S., and Hosterman, J.W.: Chemical and Mineralogical analysis of Devonian Black-shale samples from Martin County, Kentucky; Carroll and Washington Counties, Ohio; Wise County, Virginia; and Overton County, Tennessee, U.S.A. Chem. Geol. 37, 239-264 (1982).

Michel, J.: Redistribution of uranium and thorium series isotopes during isovolumentric weathering of granite. Geochim. Cosmochim. Acta 48, 1249-1255 (1984).

Minor, M.M., Hensley, W.K., Denton, M.M., and Garcia, S.R.: An automated activation analysis system. J. Radioanalytical Chem. 70, 459-471 (1982).

Pearce, J.A., Harris, N.B.W, and Tindle, A.G.: Trace element discrimination diagrams for tectonic interpretations of granitic rocks. J. Petrol. 25, 956-983 (1984).

Pettijohn, F.J.: Sedimentary Rocks. New York: Harper and Brothers 1949.

Piper, D.Z.: Rare earth elements in the sedimentary cycle: A summary. Chem. Geol. 14, 285-304 (1974).

Quinby-Hunt, M.S., McLaughlin, R.D. and Quintanilha, A.T.: Instrumentation for Environmental Monitoring, vol. 2, WATER, 2nd ed. A.E. Greenberg and G.A. Morton, Eds. New York: Wiley 1986.

Quinby-Hunt, M.S., Wilde, P., Berry, W.B.N., and Orth, C.J.: The redox-related facies of black shales. Annual Meeting of the Geological Society of America, Denver, October, 1988, 20, 193 (1988).

Quinby-Hunt, M.S., Wilde, P., and Berry, W.B.N.: Elemental geochemistry of low-calcic black shales - statistical comparison with other shales. In: Metalliferous Black Shales and Related Ore Deposits, U.S.Geological Survey Circular 1037, R.I. Grauch and J.S. Leventhal, Eds., pp. 8-15. Washington, DC: US Government Printing Office 1989.

Raiswell, R. and Berner, R.A.: Pyrite and organic matter in Phanerozoic normal marine shales. Geochim. Cosmochim. Acta 50, 1967-1976 (1986).

Schwalbach, J.R. and Gorsline, D.S.: Holocene sediment budgets for the basins of the California continental borderland. J. Sed. Petrology 55, 829-842 (1985).

Scotese, C.R. and McKerrow, W.S.: Revised World maps and introduction. In: Paleozoic Biogeography and Palaeogeography, W.S. McKerrow and C.R. Scotese, Eds., pp.1-21. London:Geological Society of London 1990.

Sholkovitz, E.R.: Artifacts associated with the chemical leaching of sediments for rare earth elements. Chem. Geol. 77, 47-51 (1989).

Stanley, J.K., Jr. and Byrne, R.H.: The influence of solution chemistry on REE uptake by Ulva lactuca L. in seawater. Geochim. Cosmochim. Acta 54, 1587-1595 (1990).

Swanson, V.: Geology and geochemistry of uranium in marine black shales. A review. Geol. Surv. Am. Prof. Paper 356-C. 220 pp. Washington, DC: U.S. Government Printing Office 1961.

Taylor, S.R. and McLennan, S.M.: The Continental Crust: Its Composition and Evolution. Oxford: Blackwell Scientific Publication 1985.

Thompson, R.N., Morrison, M.A., Mattey, D.P., Dickin, A.P., and Moorbath, S.: An assessment of the Th-Hf-Ta diagram as a discriminant for tectonomagmatic classifications and in the detection of crustal contamination of magmas. Earth Planet. Sci. Letts. 50, 1-10 (1980).

Thornton, S.E.: Basin model for hemipelagic sedimentation in a tectonically active continental margin: Santa Monica Basin, California Continental Borderland: fine-grained sediments. In: Deep-Water Processes and Facies, D.A.V. Stow and D.J.W. Piper, Eds., pp. 377-397. Oxford: Blackwell Scientific 1984.

Tlig, S. and Steinberg, M.: Distribution of rare-earth elements (REE) in size fractions of recent sediments of the Indian ocean. Chem. Geol. 37, 317-333 (1982).

Tourtelot, E.B.: Selected annotated bibliography of minor-elements content of marine black shales and related sedimentary rocks. Bull. U.S. Geol. Surv. 1293, 1-118 (1970).

Turekian, K.K. and Wedepohl, K.H.: Distribution of the elements in some major units of the Earth's crust. Geol. Soc. Am. Bull. 72, 175-191 (1961).

Vine, J.D. and Tourtelot, E.B.: Geochemistry of black shale deposits - a summary report. Econ. Geol. 65, 253-272 (1970).

Wilde, P., Berry, W.B.N., Quinby-Hunt, M.S., Orth, C.J., Quintana, L.R. and Gilmore, J.S.: Iridium abundances across the Ordovician-Silurian stratotype. Science 233, 339-341 (1986).

Wilde, P., Quinby-Hunt, M.S.Q., Berry, W.B.N., and Orth, C.J.: Palaeo-Oceanography and bio-geography in the Tremadoc (Ordovician) Iapetus Ocean and the origin of the chemostratigraphy of Dictyonema flabelliforme black shales. Geol. Mag. 126, 19-27 (1989).

Winchester, J.A. and Floyd, P.A.: Geochemical discrimination of different magma series and their differentiation products using immobile elements. Chem. Geol. 20, 325-343 (1977).

Wood, D.A.: The application of a Th-Hf-Ta Diagram to problems of tectonomagmatic classification and to establishing the nature of crustal contamination of basaltic lavas of the British Tertiary volcanic province. Earth Planet. Sci. Letts. 50, 11-30 (1980).

Wood, D.A., Joron, J.-L., and Treuil, M.: A re-appraisal of the use of trace elements to classify and discriminate between magma series erupted in different tectonic settings. Earth Planet. Sci. Letts. 45, 326-336 (1979).