Calcite isocrinid ossicles from the Middle Jurassic (Bathonian) clays in Gnaszyn (central Poland) show perfectly preserved micro- and nanostructural details typical of diagenetically unaltered echinoderm skeleton. Stereom pores are filled with ferroan calcite cements that sealed off the skeleton from diagenetic fluids and prevented structural and geochemical alteration. In contrast with high-Mg calcite skeleton of modern, tropical echinoderms, the fossil crinoid ossicles from Gnaszyn contain only 5.0–5.3 mole% Of MgCO3. This low Mg content can be a result of either a low temperature environment (ca. 10°C) and/or low Mg/Ca seawater ratio. Both conditions have been proposed for the Middle Jurassic marine environment. Occurrence of Mg-enriched central region of stereom bars of Jurassic columnal ossicle of Chariocrinus andreae is consistent with the concept of magnesium ions involvement in earliest growth phases of calcium carbonate biominerals.
Introduction
Representatives of all echinoderm clades i.e., echinoids, holothuroids, ophiuroids, asteroids, crinoids, and a few other extinct groups, form elaborate calcitic (polymorph of calcium carbonate) skeletons composed of numerous ossicles (or plates). Each plate consists of a three-dimensional meshwork of mineral trabeculae called stereom. Stereom is considered an echinoderm synapomorphy and is recognized already in Cambrian stylophorans, the most basal echinoderm clade (Bottjer et al. 2006; Clausen and Smith 2008). Individual skeletal plates behave as single calcite crystals as shown by X-ray diffraction and polarizing microscopy (Donnay and Pawson 1969), however their physico-chemical properties differ significantly from the properties of geologic or synthetic calcites. For example, echinoderm bio-calcite does not show cleavage planes typical of calcite but reveals conchoidal fracture surfaces that reduce the brittleness of the material by dissipating strain energy and deflecting crack propagation (Berman et al. 1988). The unique properties of echinoderm bio-calcite result from intimate involvement of organic molecules in the biomineralization process and their incorporation into the crystal structure. Organic components, mainly various proteins and glycoproteins, constitute 0.10– 0.26 wt% of modern echinoderm bio-calcite (Weiner 1985; Wilt 1999) and are incorporated in the skeleton at different structural levels. At the nanoscale, a distinct feature of calcium carbonate biomineral is its composite structure: mineral grains (commonly 30–100 nm in diameter) are closely associated with an organic material. Nanocomposite structure (nanograins 20–100 nm) have also been reported for bio-calcite of echinoderms, i.e., spines of extant echinoids (Oaki and Imai 2006). Fossil calcium carbonate biominerals that are not diagenetically altered also preserve nanocomposite structure (Stolarski and Mazur 2005).
The goal of this work is to describe fine scale structural and geochemical features of fossil crinoid ossicles recovered from the Middle Jurassic (ca. 167 Ma) sediments in Gnaszyn, Poland. These clay deposits are known for their excellent preservation of aragonitic skeletons that are not dissolved or recrystallized as in many other types of sedimentary rocks of the same age. As aragonite is metastable at ambient temperatures and pressures, and easily transforms into calcite, which is a stable CaCO3 polymorph, it can only be preserved if sealed off from pore-water penetration. Preservation of aragonite fossils at Gnaszyn suggests that crinoid bio-calcite might also preserve original structural and geochemical signatures. Note that modern crinoids form magnesium-rich calcite that is also susceptible to diagenesis.
Materials and methods
Materials.—The fossil crinoids were collected from an active clay pit at Gnaszyn Dolny (abbreviated “Gnaszyn”) approximately 10 km southwest of Częstochowa (Fig. 1) in Polish Jura Chain of south-central Poland. The sedimentary rocks exposed at Gnaszyn are represented mainly by dark-grey calcareous silty clays, occasionally with levels of sideritic concretions and are dated as Middle Bathonian (Tulites subcontractus and Morrisiceras morrisi Chrones) and Upper Bathonian (Procerites hodsoni Chron = Bremeri and early Prohecticoceras retrocostatum Chron in Submediterranean zonation); see Majewski 1997; Zatoń et al. 2006; Matyja and Wierzbowski 2000, 2006. Both the clays and the sideritic concretions are rich in fossils that include bivalves, gastropods, ammonites with preserved aragonitic skeleton as well as belemnites, crustaceans, and echinoderms (Gedl et al. 2003, 2006). Columnals and pluricolumnals of crinoids are the most frequent echinoderm remains at this locality (Salamon and Zatoń 2007). Many of them are dark (black) in color whereas light (beige) columnals represent ca. 10% of the collection. Two columnals from the Middle Bathonian clays of Gnaszyn were examined in more detail: (i) stellate and beige in color columnal of Chariocrinus andreae (Desor, 1845) (Fig. 2A), and (ii) circular and dark columnal of Balanocrinus berchteni Hess and Pugin, 1983 (Fig. 2B).
Methods.—Atomic Force Microscopy (AFM) was performed with a MultiMode Nanoscope IIIa (Digital Instruments, Veeco), following procedures described in Stolarski and Mazur (2005). Standard silicone nitride cantilevers were used for measurements in contact mode. The polished samples were then etched in 1% ammonium persulfate in McIlvain buffer (pH = 8) for 10 min., followed by rinsing in deionized water and drying. Trace element analyses were performed with the Cameca NanoSIMS N50 at the Muséum National d'Histoire Naturelle in Paris, following procedures described in Meibom et al. (2004, 2008). The samples were polished to 0.25 µm diamond finish then Au-coated. Using a primary beam of O-, secondary ions of 24Mg+, 44Ca+, and 88Sr+ were sputtered from the sample surface and detected simultaneously (multicollection-mode) in electron-multipliers at a mass-resolving power of ∼4500 (M/DM). At this mass-resolving power, the measured secondary ions are resolved from potentially interferences. Data were obtained from a pre-sputtered surface in a series of rasters with the primary ions focused to a spot-size of ∼200 nm. The primary beam was stepped across the sample surface with a similar step-size. The measured 24Mg/44Ca and 88Sr/44Ca ratios were calibrated against analyses of carbonate standards of known composition and against spot elemental analyses performed with a scanning microscope Philips XL-20 coupled with the EDS detector ECON 6 (system EDX-DX4i) at the Institute of Paleobiology, Warsaw, Poland. The BSE detector of the Philips XL-20 microscope was used for polished and carbon-coated specimens, allowing to distinguish between material of lower vs. higher molecular density.
Results
Microscale.—In both columnals, the stereom framework is well preserved. The columnal of Balanocrinus berchteni Hess and Pugin, 1983 has a circular outline and relatively large, triangular petals. They are encircled by 6–8 marginal crenulae: marginal crenulae are of the same size but distinctly thicker than adradial ones. In the star-shaped columnais of Chariocrinus andreae (Desor, 1845), lancet-shaped petals are encircled by well developed crenularium with 12–16 crenulae associate with each petal. In transverse sections of both columnals two distinct regions can be distinguished: (i) petaloid regions built of regular meshwork of galleried type stereom (α-stereom of Roux 1975) and (ii) interpetaloid areas composed of irregular, labyrinthic stereom (β-stereom of Roux 1975). Interstereom spaces in both specimens are infilled with secondary deposits, mostly of ferroan calcite. Secondary deposits developed in the peripheral zone of the B. berchteni plate contain numerous framboidal pyrite grains, some of them attaining 5–8 µm in diameter (Fig. 2B1, B2, B6–B8) that dyes this columnal black. Conversely, no pyrite accumulations were observed in interstereom ferroan calcite infilling of light colored plate of C. andreae.
Nanoscale.—Calcite stereom skeleton of both columnals consists of nanograins ca. 100 nm in diameter. No distinct variation in the nano-texture was observed in the examined samples. Individual grains have a semicircular outline and are separated from each other by spaces of a few nanometers width (Fig. 2A4, A5, A9, A10, B4, B5, B10, B11). Occasionally, larger, up to 300 nm, rounded aggregates of smaller nanograins are visible (Fig. 2A9, A10). There is a sharp boundary in nanoscale texture between the stereom (distinct nanograins) and interstereom deposits (relatively smooth surface with low value of roughness factor). The stereom surface has deeper grinding scratches and its relief is lower than that of adjacent interstereom deposits that suggests it has a lower hardness.
Geochemistry.—BSE mode images show clear contrast between chemical composition of stereom and interstereom sediment. BSE enhances the atomic number contrast and elements with lower atomic numbers appear darker, whereas those with higher atomic numbers, e.g., iron appear brighter. There is some diffuse heterogeneity in BSE images of individual stereom trabeculae with a slightly darker region in the central part and lighter at the edge. The Mg composition of ossicles was averaged over 10 spot EDS analyses. In Chariocrinus andreae the Mg% range was 1.0–1.4 (average = 1.2 ± 0.13) which corresponds to ca. 5.0 mole% of MgCO3. The concentration of Mg in Balanocrinus berchteni plate was slightly higher: 0.9–1.7 (average = 1.3 ± 0.22), which corresponds to ca. 5.3 mole% MgCO3. Further evidence of the effects observed in BSE mode was provided by NanoSIMS elemental mappings and line scans taken from different regions of C. andreae plate (the B. berchteni sample was affected by strong charging thus is not shown here). The main feature of the Mg mapping and profile (Fig. 3A–C) is low Mg concentration in interstereom deposits (Mg/Ca 20–30 mmol/mol) and higher concentration in the stereom (Mg/Ca 60–70 mmol/mol) with a peak in the middle part of the trabecular bar (Mg/Ca ca. 80 mmol/ mol). Sr concentration (Fig. 3D–F) is generally very low but similarly to the Mg distribution it is lower in interstereom deposits (Sr/Ca 0.2–0.4 mmol/mol) and higher in the stereom (Sr/Ca 1.0–1.2 mmol/mol). Due to low count rate, any possible differences in Sr distribution within the stereom bar cannot be resolved.
Discussion
Complex macro- and micromorphology of the echinoderm ossicles results from precisely orchestrated biomineralization processes. Echinoderm ossicles are produced inside a syncytial membrane formed by several cells (e.g., Okazaki 1960) and a microporous structure of the stereom results directly from the 3D organization of the organic lining. There is growing evidence about a direct link between biochemical properties of some protein and glycoprotein components and modulation of the growth of the echinoderm skeleton (Ameye et al. 2001; see also review in Bottjer et al. 2006). Organic components that are present in the crystallization medium are incorporated into calcite crystals which results in their nanoporous internal structure (Robach et al. 2005, see also biomimetic experiments of Park and Meldrum 2004; Lu et al. 2005; Li and Estroff 2007). There is also growing evidence that nanocomposite structure, with nanograins 20–100 nm in size, is a universal organization of calcium carbonate biocrystals (Dauphin 2001; Cuif et al. 2004; Rousseau et al. 2005; Stolarski and Mazur 2005) and that precursor of the nanograins are amorphous calcium carbonate (ACC) granules containing high level of inorganic and organic components (Politi et al. 2004). Oaki and Imai (2006) observed the nanocomposite structure (with “nanobricks” 20–40 nm in size) in all examined echinoid stereom samples with FESEM and FETEM. Nanocomposite structure of the stereom is also supported by staining experiments with different types of organic dyes homogenously dispersed within the nanoscale architecture of the echinoid spine stereom (Oaki and Imai 2006). Examined fossil isocrinid ossicles display all micro-(stereom) and nanoscale (nanograins) details mat are observed in Recent echinoderm bio-calcite hence appear not markedly affected by diagenesis. This is explained by their preservation in clay sediments and filling of their stereom pores with ferroan calcite cement that sealed off the skeleton from digenetic fluids and prevented its structural and geochemical alteration (Dickson 2004). Lack of diagenetic transformation is also supported by co-occurrence of originally calcitic (stable CaCO3 polymorph) and aragonitic (metastable CaCO3 polymorph) fossils in surrounding clay layers. Echinoderm ossicles from Gnaszyn seem therefore particularly suitable for geochemical studies.
Extant echinoderms produce calcite skeleton that usually contains high amount of Mg. However, it is also known that Mg content varies significantly (5–19 mol% MgCO3) with physiological/environmental factors. Different skeletal parts of a single organism may contain different amount of Mg. For different taxa these values may vary as well and overall Mg content in the skeleton seems to be correlated with seawater temperature (Clarke and Wheeler 1922; Chave 1954; Weber 1969, 1973; Roux et al. 1995; Borzęcka-Prokop et al. 2007; David 1998). On the other hand, Dickson (2002, 2004) argued that Mg2+ composition of fossil echinoderm ossicles from “paleotropical or paleosubtropical” settings (such sample selection was made to minimize temperature effects) reflects secular changes of the seawater Mg/Ca geochemistry during the Phanerozoic. For example, Jurassic ossicles were preserved as Mg calcite with low, 4–6 mole% MgCO3 content, whereas Pennsylvanian ossicles had high, 9–12 mole% MgCO3 content and were preserved as calcite/microdolomite (Dickson 2004). This difference perfectly matches change between high Mg/Ca ratio of Pennsylvanian “aragonitic sea” and low Mg/Ca ratio of the Jurassic “calcitic sea”.
At a glance, magnesium content in the Middle Jurassic crinoid ossicles from Gnaszyn (5.0–5.3 mole% of MgCO3) match perfectly estimated effects of echinoderm Mg incorporation in low (about 1.4) Mg/Ca seawater ratio settings (Dickson 2004). However, Gnaszyn crinoids come from sediments deposited in what were probably colder settings. Based on δ18O data from the calcitic belemnite rostra (possibly nectobenthic life style), Marynowski et al. (2007) and Wierzbowski and Joachimski (2007) suggested mean palaeotemperatures of 13.1°C or even 9.2°C, whereas δ18O data from benthic oysters and trigoniid bivalves suggest 10.1°C and 7.4°C values, respectively. Consequently, the low magnesium content in Gnaszyn ossicles may not only be due to low Mg/Ca ratio of Middle Jurassic seawater but also due to low seawater temperatures. In fact, Dickson's (2002, 2004) conclusions regarding paleotropical or paleosubtropical settings with “temperatures probably between 20°C and 30°C” were indirect1 and need to be verified with aid of similar stable isotopic tools. In future work we will try to address these interpretative problems using similarly well preserved crinoids from Poland and elsewhere with special focus on possible “vital effects” in Mg2+ incorporation and distribution between different parts of isocrinid skeleton that is known to exist in modern crinoids (David 1998).
Recent, fine scale geobiochemical mappings of echinoderm skeleton confirm that Mg is distributed heterogeneously in the stereom, reflecting strict biological biomineralization control. Robach et al. (2006: 92) found that large differences in Mg concentration in the teeth of sea urchins coincide with the distribution of aspartic acid (Asp) suggesting that macromolecules enriched in Asp control formation of very high Mg calcite. Such sub-micrometer scale biomineralization control is suggested also for the crinoid skeleton. Dickson (2004: fig. 1B, F) observed diffuse banding in stereom Mg calcite, generally with individual bands parallel to the stereom surface and suggested that “banding is similar to the concentric pattern of organic laminae (…) in modern echinoids” (Dickson 2004: 356). Our preliminary NanoSIMS mapping of the ossicle of C. andreae (Fig. 3A,B) shows that inner region of the stereom bar is Mg-enriched. Although the earliest phases of crinoid stereom growth have not been documented (most studies focus on anatomical and early phases of crinoid plate development, e.g., Lahaye and Jangoux 1987; Haig and Rouse in press) they probably can be compared with stereom growth in other echinoderms. In echinoids, the inner core of spine trabeculae is preferentially enriched in N-glycoproteins that are assumed to stabilize an amorphous calcium carbonate (ACC) phase (see Ameye et al. 2001; review in Wilt and Ettensohn 2007). The central organic-enriched region is also similar to that described in larval echinoid spiculae (Benson et al. 1983) whose precursor phase is also ACC (Beniash et al. 1997). Surveys of the earliest growth phases of various carbonate invertebrate skeletons consistently show a link between organic and magnesium enriched first-formed regions and formation of precursory ACC skeleton phase (Aizenberg et al. 2002; Han et al. 2005; Weiner et. al. 2005; Meibom et al. 2008). We suggest herein, that Mg-enrichment in the central region of stereom bar of Jurassic C. andreae and diffuse stereom banding observed by Dickson (2004) may represent remains of original trace element signatures of transient ACC (compare Meibom et al. 2004 for scleractinian coral Mg-enriched banding). Further comparative studies with Recent material are planned to test this hypothesis.
Acknowledgements
Special thanks are due to Mariusz A. Salamon and Michał Zatoń (Faculty of Earth Sciences, University of Silesia, Poland) for access to material and for providing literature data on stratigraphy of Gnaszyn locality. Tomasz Baumiller (Museum of Paleontology, University of Michigan, USA) and Andrew B. Smith (Department of Palaeontology, The Natural History Museum, London) are thanked for constructive criticism, which helped improve the manuscript. The work was carried out with the financial support of Polish Ministry of Science and Higher Education, project no. N307-015733, Department of Chemistry, University of Warsaw, project BW-120000-501/68-179210, and Museum National d'Histoire Naturelle and the Agence National de la Recherche.