Stable carbon and oxygen isotope compositions of extant crinoidal echinoderm skeletons

Stable carbon and oxygen isotope compositions of extant crinoidal echinoderm skeletons

Chemical Geology 291 (2012) 132–140 Contents lists available at SciVerse ScienceDirect Chemical Geology journal homepage: www.elsevier.com/locate/ch...

1MB Sizes 0 Downloads 0 Views

Chemical Geology 291 (2012) 132–140

Contents lists available at SciVerse ScienceDirect

Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo

Research papers

Stable carbon and oxygen isotope compositions of extant crinoidal echinoderm skeletons Przemysław Gorzelak a,⁎, Jarosław Stolarski a, Krzysztof Małkowski a, Anders Meibom b a b

Institute of Paleobiology, Laboratory of Biostructures and Biomineralization, Polish Academy of Sciences, Twarda 51/55, PL-00-818 Warsaw, Poland Muséum National d'Histoire Naturelle, Laboratoire d'Etude de la Matiere Extraterrestre, USM 0205 (LEME), Case Postale 52, 61 rue Buffon, 75005 Paris, France

a r t i c l e

i n f o

Article history: Received 19 October 2010 Received in revised form 11 October 2011 Accepted 11 October 2011 Available online 20 October 2011 Editor: U. Brand Keywords: Vital effects Light stable isotopes Echinoderms Biomineral

a b s t r a c t The variability of δ13C and δ18O was determined within the columnal facet of individual ossicles, within different regions of skeletons and within bulk skeletons of extant stalked crinoids. Isotopic compositions of individual ossicles may vary by up ~ 1‰ for both isotopes, whereas isotopic variability within a skeleton may be as high as ~2.8‰ for δ13C and ~1.2‰ for δ18O. In contrast, mean isotopic compositions and variations are similar for different specimens of a single species from any particular locality. Isotopic variation was evaluated between higher taxonomic groups of crinoids, including Isocrinida, Comatulida, Bourgueticrinida and Cyrtocrinida. Skeletons of isocrinids, comatulids and bourgueticrinids are consistently more negative in δ13C than those of cyrtocrinids. This difference may be as high as ~10‰, and is unrelated to the place of origin. Such isotopic differences reflect distinct physiological differences between crinoid groups we studied. Overall, their δ18O values show weak temperature dependence, which are overshadowed by the strong influence of physiological or vital effects on the isotopic composition of crinoid skeletal carbonate. Thus great caution needs to be exercised when using the stable isotope composition of crinoids as an environmental proxy. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Carbon (δ13C) and oxygen (δ18O) isotope compositions of extant echinoderm skeletons have been studied by several authors and from different perspectives. Weber and Raup (1966, 1968) and Weber (1968) analyzed the bulk isotopic compositions of representatives of nearly all echinoderm clades, with special emphasis on echinoids (see also Richter and Bruckschen, 1998). They found that the mechanism of carbon and oxygen isotope fractionation in echinoderms is very complex, and depends on both environmental and biological (vital) factors. In these studies, they reported substantial differences between species, as well as between higher-level taxonomic units. They also found variations within different ossicles of single individual. Weber (1968) observed that, at a given precipitation temperature, the skeletons of crinoids are enriched in 12C and moderately enriched in 16O with respect to calcium carbonate that was inorganically precipitated from seawater. Further work by Oji (1989) suggested that δ18O variations in the stalk of Metacrinus rotundus (Carpenter), a stalked crinoid with a very shallow habitat, reflect annual changes in water temperature. Later, isotopic analyses of pentacrinid stalked crinoids indicated that the δ18O of their skeletons is depth dependent (David, 1998) while δ13C variations are related to the metabolic rate and/or reflect taxonomical groupings (Roux et al., 1995; David, 1998). Baumiller (2001) subsequently suggested that

⁎ Corresponding author. Tel.: + 48 22 6978893; fax: +48 22 6206225. E-mail address: [email protected] (P. Gorzelak). 0009-2541/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.chemgeo.2011.10.014

crinoid skeletons are of limited paleoenvironmental use due to strong “vital” effects on their stable isotope composition. However, he also concluded that crinoid stable isotopes can be applied for various (paleo)biological purposes, such as the identification of autotomy events (regenerated parts of arms following self-amputation), and the reconstruction of the soft tissue distribution (muscles vs. ligament) within skeletal elements (see Baumiller (2001: 111)). The main aim of the present paper is to review, complement and expand previous work on light stable isotope geochemistry of extant stalked crinoids. In particular, our goal is to provide insights into: (1) isotopic variability within the columnal facet of individual ossicles, (2) isotopic variability within a skeleton of a single animal, (3) isotopic variability between different specimens of a single species, (4) differences between the isotopic variability observed among higher taxonomic groups, and (5) potential relationships between δ 18O and parameters (such as temperature) of the ambient environment. 2. Materials and methods Skeletons of four isocrinid crinoid species (Isocrinida) collected by submersible and fishery boats were selected for this study (Table 1). The specimens of Neocrinus decorus (Thomson) and Endoxocrinus parrae parrae (Gervais) came from the North Atlantic Ocean on the southwestern margin of Little Bahama Bank (the west end of Grand Bahama Island); depths ranged from 402 m to 425 m. There are weak currents and little temperature variability at this site (Leaman et al., 1987; Baumiller, 2001). The overlying waters are oligotrophic.

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

133

Table 1 Taxonomy, localities, co-ordinates, collection depths and sea-water temperature of crinoid specimens investigated in this study. Crinoid taxa investigated in the present study

Locality

Co-ordinates

Depth [m]

Temperature [°C]

Metacrininae, Isocrinida Metacrinus rotundus Metacrinus rotundus Metacrinus rotundus

NE Suruga Bay, southern coast of central Japan NE Suruga Bay, southern coast of central Japan NE Suruga Bay, southern coast of central Japan

35°03′N, 138°48′E 35°03′N, 138°48′E 35°03′N, 138°48′E

140 140 140

15 15 15

Diplocrininae, Isocrinida Endoxocrinus parrae parrae

North Atlantic Ocean, Grand Bahama Island

26°37.7′N, 78°58.7′W

402

16.5

North Atlantic Ocean, Grand Bahama Island North Pacific Ocean, Shima-Spur, off Kii Peninsula, Honshu Island, Japan

26°37.6′N, 78°58.7′W 34°00.83′N, 136°53.79′E–34°01.42′N, 136°51.80′E

425 805–852

16.5 –

Holopodidae, Cyrtocrinida Cyathidium foresti

North Atlantic Ocean, Pico Ridge near Azores

750–590

12a

Cyathidium foresti

North Atlantic Ocean, Pico Ridge near Azores

750–590

12a

Cyathidium foresti

North Atlantic Ocean, Pico Ridge near Azores

38°07.866′N, 27°34.223′W–38°08.037′N, 27°34.172′W 38°07.866′N, 27°34.223′W–38°08.037′N, 27°34.172′W 38°07.866′N, 27°34.223′W–38°08.037′N, 27°34.172′W

750–590

12a





Balanocrininae, Isocrinida Neocrinus decorus Hypalocrinus naresianus

Caledonicrininae, Order uncertain Caledonicrinus vaubani Southwestern Pacific Ocean (no precise locality data) a



Approximated from Wisshak et al., 2009.

The local sea floor is sloping. It consists of submarine cemented hardgrounds covered with sediment composed almost entirely of skeletal and lithic grains (e.g. Messing et al., 2007). The specimens of M. rotundus (Carpenter) were collected by a fishing boat at 140 m depth in NE Suruga Bay, on the southern coast of central Japan. The bottom topography of this site is a steep slope near the edge of the continental shelf. The hydrological circulation within the bay is influenced by the inflow of both oceanic water (the Kuroshio current) and fluvial discharge from the Oi, Abe, Fuji, and Kano Rivers (Ohta, 1983). The steepness of the bay facilitates access to the deep basin from nearby land (Takahashi et al., 1997). The annual range of water temperature here is several degrees whereas the annual fluctuation of salinity is usually less than 0.3‰ (Oji, 1989). The specimen of Hypalocrinus naresianus (Carpenter) was obtained from trawling at Shima Spur, off Kii Peninsula (start-depth 805 m, end-depth 852 m). The sea floor at this site is a flat plateau covered by loosely consolidated blue-gray silt (more details in Kitazawa et al., 2007). Three cyrtocrinid specimens of Cyathidium foresti (Cherbonnier and Guille) collected from the dredge at station POS286-203-7 at the Pico Ridge (near Azores, eastern Atlantic) from 750 m to 590 m water depth were also analyzed (Table 1). The Pico Ridge is a submarine basalt flow, 85 km long (e.g. Beier, 2006). This is stable with respect to temperature and salinity. Judging from data collected from the southern Faial Channel in the Azores Archipelago, that is located near the collection site (Wisshak et al., 2009), there are likely small seasonal fluctuations. Finally, a specimen of Caledonicrinus vaubani (Avocat and Roux) from the Southwestern Pacific Ocean (no precise locality data) was analyzed. A typical crinoid consists of two basic sections; a segmented stalk and a crown (Fig. 1). Their skeleton is made up of numerous calcareous plates (ossicles) held together by ligaments and in some cases muscles. All specimens investigated in the present study were disarticulated and separated from the organic tissues by soaking in 5% sodium hypochlorite for 24 h. Then, the ossicles were washed and dried at room temperature. For the microscale analysis, the individual ossicles were drilled with a diamond-tipped micro-drill to produce a fine powder. For the intraindividual and intra-specific comparisons, we powdered each ossicle.

We then analyzed the homogenized microsamples. Isotopic reproducibility of the powders was within the limits of analytical error. The measurements were carried out at the Institute of Paleobiology of the Polish Academy of Sciences in Warsaw. Carbonate powders were reacted with 100% phosphoric acid at 75 °C using a KIEL IV online automatic carbonate preparation line connected to Finnigan Mat delta plus mass-spectrometer. All isotopic data were reported in permil deviation relative to the VPDB reference via the NBS 19 standard. The precision (reproducibility of replicate analyses) of both carbon and oxygen isotope analysis was usually better than ±0.15‰. In the text, ‘big delta’ notation, such as Δ 13C and Δ 18O designates the total range (i.e. the most positive minus the most negative) in isotopic composition within a comparable set of data.

Fig. 1. Reconstruction of stalked crinoid: a) showing examined skeletal parts/types of ossicles: (1) distal brachial, (2) axillary brachial, (3) basal, (4) columal b): close-up of columnal facet.

134

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

Fig. 2. Isotopic variability within a single ossicle (a–c).

3. Results 3.1. Variation within a columnal facet of a single ossicle To determine whether substantial isotopic differences exist within a columnal facet of a single ossicle, we analyzed several microsamples

from the proximal nodal columnals of 3 specimens (Fig. 2a-c). Separate analysis of the petaloid and interpetaloid areas of the same columnal facet of a single ossicle revealed variations up to ~1‰ in both δ 13C and δ 18O (Table 2). Both isotopes are distributed heterogeneously within the columnal facet without any pattern. Isotopic signatures overlap between simultaneously formed stereom of petaloid

Table 2 Isotope values for specimens investigated. The STD is the standard deviation in permil among the N samples for each dataset. Skeletal part

Mean δ13C

STD

Δ13C

Mean δ18O

STD

Δ18O

N

Columnal facet Crown Stalk Whole specimen Crown Stalk Whole specimen Crown Stalk Whole specimen

− 6.29 − 5.07 − 6.67 −5.87a − 5.35 − 6.68 − 6.02a − 5.23 − 6.47 − 5.85a

0.51 0.24 0.04 0.73 0.43 0.19 0.79 0.22 0.14 0.39

1.15 0.48 0.06 2.02 0.85 0.27 1.85 0.40 0.55 1.72

−1.33 −1 − 1.58 − 1.29a − 1.19 − 1.59 − 1.39a − 0.82 − 1.11 − 0.97a

0.35 0.28 0.07 0.35 0.17 0.2 0.27 0.21 0.29 0.29

0.94 0.52 0.1 1.2 0.35 0.29 0.72 0.42 1.06 1.06

6 3 2 11 3 2 5 3 31 34

Columnal facet Crown Stalk Whole specimen

− 6.01 − 4.11 −6.32 − 5.22a

0.18 0.40 0.05 1.09

0.4 1 0.6 2.81

− 1.2 −0.59 − 1.06 − 0.83a

0.11 0.24 0.20 0.34

0.26 0.57 0.39 0.94

4 5 2 11

Crown Stalk Whole specimen Columnal facet Crown Stalk Whole specimen

− 6.43 − 7.66 −7.05a − 6.9 − 5.86 −6.9 − 6.38a

0.52 0.23 0.71 0.17 0.23 0.16 0.53

1 0.6 2.08 0.44 0.42 0.44 1.5

− 0.71 −1.74 − 1.23a + 0.31 + 0.77 + 0.31 + 0.54a

0.2 0.09 0.55 0.09 0.12 0.09 0.24

0.4 0.26 0.94 0.25 0.24 0.25 0.71

3 5 8 6 3 7 16

Cyrtocrinida Holopodidae Cyathidium foresti (specimen no. 1) Cyathidium foresti (specimen no. 2) Cyathidium foresti (specimen no. 3)

Whole specimen Whole specimen Whole specimen

+ 2.23 + 2.24 − 0.37

0.12 0.1 0.36

0.22 0.15 0.51

+ 2.9 + 2.77 + 1.76

0.13 0.02 0.09

0.26 0.03 0.13

3 2 2

Order uncertain Caledonicrininae Caledonicrinus vaubani

Stalk

+ 1.87

0.08

0.19

+ 2.24

0.25

0.61

5

Crinoid taxa Isocrinida Metacrininae Metacrinus rotundus (specimen no. 1)

Metacrinus rotundus (specimen no. 2)

Metacrinus rotundus (specimen no. 3)

Diplocrininae Endoxocrinus parrae parrae

Balanocrininae Neocrinus decorus

Hypalocrinus naresianus

a

Mean values represent average values between mean crown and stalk values.

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

Fig. 3. Range of isotopic values for different morphological parts of a single specimen: a) Endoxocrinus parrae parrae b): Hypalocrinus naresianus c): Neocrinus decorus.

135

136

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

Fig. 4. Range of isotopic values for different morphological parts of a single specimen: a–c) Metacrinus rotundus.

areas and the stereom of interpetaloid areas of the same facet (Fig. 2a–c). 3.2. Variation within the skeleton of a single animal In Table 2, we report isotope variations of δ 13C and δ 18O within six specimens of four species (c.f. Figs. 3 and 4). In order to minimize the effect of within-ossicle isotopic variation on the stalk-crown data sets, we powdered and homogenized the samples before analysis. Without exception, the average δ 13C of the crinoid stalk was isotopically more negative than that of the crown. δ 13C differences between stalk and crown range from 1.0‰ in H. naresianus to 2.2‰ in E. parrae parrae (Table 2). These stalk-crown differences in δ 13C are statistically significant; t-test p-values b 0.05. For δ18O the stalk–crown differences were less pronounced, ranging from 0.3‰ in M. rotundus (specimen no. 3, Fig. 4a) to about 1‰ in N. decorus (Table 2). The average δ18O composition of the stalk was found to be systematically more negative than that in the crown. In one case (M. rotundus, specimen no. 3, Fig. 4a), however, we also observed substantial δ18O variability within the stalk, with Δ18O up to about 1‰. T-tests showed that the observed stalk–crown differences in δ18O were statistically significant (p-valuesb 0.05) only for H. naresianus and N. decorus.

3.3. Variation between different specimens of a single species The overall variation in δ 13C and δ 18O between individuals of M. rotundus for a given locality is small. Observed isotopic differences are close to analytical error (Δ 13C up to 0.17‰ and Δ 18O up to 0.42‰) with calculated standard deviations of 0.09 and 0.22‰ for δ 13C and δ 18O, respectively. For C. foresti the isotopic variations are larger (Δ 13C up to 2.6‰ with standard deviation 1.5‰, and Δ 18O up to 1.1‰ with standard deviation 0.62‰)). 3.4. Variation at higher taxonomic levels δ 13C compositions within isocrinid families appear to be similar (i.e., there are smaller variations at lower taxonomic levels). However, differences become larger at higher taxonomic levels. Fig. 5 combines bulk δ13C compositions from this study with literature values for animals living under comparable environmental conditions. Average δ13C values for cyrtocrinids fall in the range ~−1 to ~+2‰ and caledonicrinids have average δ13C values between ~−2‰ and ~+2‰. In contrast, δ13C is substantially more negative, ranging between ~−4 and ~−9‰, in all samples from the three other orders: isocrinids, comatulids and bourgueticrinids (Fig. 5).

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

137

Fig. 5. The δ13C values of the five major crinoid orders (after Baumiller (2001) and our own data [in gray]). DW — deep water comatulid. Remark: (*) — the systematic position of Caledonicrinus is difficult to establish (previous cladistic morphological analysis did not link this genus with any clade but molecular analysis placed it amongst cyrtocrinids; see Cohen et al., 2004).

Differences are less pronounced in mean δ18O compositions between representatives of two major clades (i.e. isocrinids versus cyrtocrinids and caledonicrinids):−1.4 to +0.5‰ for isocrinids and +1.8 to +2.9‰ for cyrtocrinids and caledonicrinids. 3.5. Relationship between temperature and skeletal δ 18O To determine whether there is a relationship between skeletal δ18O and the mean annual temperature, we analyzed our data together with previous results where habitat temperatures were determined to a sufficient accuracy (Weber, 1968; Baumiller, 2001). As shown in Fig. 6, the relationship between the skeletal δ18O and the mean annual temperature of crinoid habitats is negatively correlated with moderate correlation coefficient. 4. Discussion 4.1. Isotopic inhomogenity of a columnal facet of single skeletal elements We have observed distinct isotopic variations within a single ossicle up to ~1‰ in both δ18O and δ13C (Fig. 2; Table 2). Such isotopic variations (δ18O) are often explained by kinetic effects linked to differences in

skeleton accretion rate. These can, in turn, be linked to differences in cellular activity/metabolic rates, and thus recorded by δ13C variations. We therefore expected to find some correlation between measured isotopic values and specific microstructural regions of the columnals. It is known that different microstructural regions of columnals are formed during complex phases of growth. These include rapid calcite accretion in peripheral directions, very slow increase in the columnal height, and development of the culmina and ridges into a radiating pattern on the columnal facet and their subsequent augmentation and the intercalation of new ones between the older ones (more details in Brower, 1974; Simms, 1989). Baumiller (2001) observed that, within a single brachial, δ13C varied by almost 0.5‰ between different stereom regions (i.e., ligament and muscle stereom). However, in our study we have not observed a clear pattern in the distribution of the light stable isotopes within columnals. Petaloid areas of the same facet, which grow at the same time, were isotopically distinct. Similar variations occurred in interpetaloid stereom of the same columnal, which also grow simultaneously. The petaloid areas were sometimes found to be isotopically more negative than interpetaloid areas of the same facet and vice versa. In other words, a particular part of the same facet did not reproduce a particular isotopic signature. We assume therefore that higher-resolution analytical studies (at the nano-scale level) are needed to explain the observed δ13C and δ18O variations within the columnal facet of single skeletal elements.

4.2. Effect of skeletal anatomy

Fig. 6. The δ18O values as a function of temperature at collection site (after Weber, 1968 [white diamonds]; Baumiller, 2001 [black squares] and our own data [gray circles]).

δ 13C and δ 18O are highly variable within the main parts of a single specimen of extant isocrinid N. decorus. For example, δ 13C and δ 18O variability between the stalk and the crown of this species may amount to as much as ~1.6‰ in δ 13C and ~ 0.8‰ in δ 18O (Baumiller, 2001). Our study of additional crinoid taxa also indicates that the stalk is isotopically more negative than the crown of the same individual. Growth rates of the crinoid crown and stalk are comparable (Vail, 1989; Messing et al., 2007), and therefore cannot account for the isotopic differences. These differences are more likely the result of different physiological activities between major regions of the crinoid skeleton. The crinoid crown contains the water vascular system used for locomotion, suspension feeding and respiration (Ausich et al., 1999). The respiration process in the crinoid crown is probably more efficient and might allow for less isotopic exchange between metabolic CO2 and seawater bicarbonate, hence a less isotopically fractionated crown. Additionally, differences in

138

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

skeletogenic cells and their local respiration may account for the observed isotopic differences. 4.3. Intraspecific vital effect Intraspecific isotopic variations for M. rotundus were found to be very small. However, isotopic variations between specimens of C. foresti are relatively high, especially in δ 13C. This “intraspecific vital effect” probably results from different respiratory modulations of δ 13C (see below). 4.4. Taxonomic and environmental effects Baumiller (2001) was the first to note that δ 13C values for comatulids (Comatulida), isocrinids (Isocrinida), and bourgueticrinids (Bourgueticrinida sensu Rasmussen, 1978) are indistinguishable from each other (max. value = ~−3.5‰, min. value = ~−9‰) as are the carbon compositions of cyrtocrinids (Cyrtocrinida) and millericrinids (Millericrinida) [max. value = ~+2‰, min. value = ~−1.5‰]. The latter author noted, however, that δ 13C differences between these two clades should be verified, because his cyrtocrinid–millericrinid specimens came from much deeper and cooler waters than the representatives of other taxa. Thus, some differences in isotopic composition ascribed to biological control could in fact result from a sea-water temperature effect. Our isotopic data for various isocrinid and cyrtocrinid taxa greatly strengthen the hypothesis of Baumiller (2001) that significant isotopic differences exist at higher taxonomic levels. In our isocrinids, δ 13C ranged from −7.9 to − 3.6‰, and δ 18O ranged from −1.9 to + 0.9‰. By contrast, the isotopic composition of cyrtocrinids is similar to the values for modern Mg-calcite marine cements (+3 to +4‰ in δ 13C and −0.5 to + 2‰ in δ 18O; see Gonzalez and Lohmann, 1985). Furthermore, the mollusk (bivalve shell) that was attached to one of our cyrtocrinid specimens also yields similar isotopic values (~+1.8‰ in δ 13C and ~+3.4 in δ 18O). Given that the shells of mollusks, despite some complexities, are generally considered as environmental recorders (e.g. McConnaughey and Gillikin, 2008), these data suggest that cyrtocrinids, are unique among crinoid taxa, that have been studied to date, in that they fractionate light stable isotopes to a small extent. One can suggest that differences in isotopic composition between representatives of higher level taxa could result from variations in the δ 13C of their food. However, this hypothesis is unlikely because our cyrtocrinids came from an environment similar to that of the other stalked crinoids (e.g. isocrinids) and their food sources were therefore probably quite similar. Moreover, despite some differences in diet in various isocrinids (including the presently analyzed genera Hypalocrinus and Metacrinus) and comatulid taxa (see Kitazawa et al., 2007), the carbon composition of their skeletons is almost the same. Additionally, it has been shown that only the skeletons of land snails, birds, and mammals produce shell and bone carbonates that isotopically reflects the animal's diet (e.g., Sullivan and Krueger, 1981; Goodfriend and Margaritz, 1987). Differences in growth rates and/or temperature-related differences are also not a likely source for the differential. There are significant differences in the growth rates among the isocrinid species we have studied. Maximum extrapolated growth rates of individual specimens of E. parrae parrae, N. decorus and M. rotundus were 3.6; 17.1 and 60 cm/year, respectively (Oji, 1989; Messing et al., 2007). Furthermore, specimens of E. parrae parrae and N. decorus were collected from deeper and cooler environment than specimens of M. rotundus. Despite these significant differences in growth rates and in temperature/depth, their isotopic compositions are quite similar. It has been argued that isotopic disequilibria often result from “kinetic” and “metabolic” effects (McConnaughey, 1989a,b). In the former case, the effect arises from discrimination against 18O and 13C isotopes during CO2 hydration and hydroxylation. The resulting

simultaneous depletions of δ 18O and δ 13C are as large as 4‰ and 10 to 15‰, respectively. This effect tends to produce nearly linear correlation between skeletal δ 18O and δ 13C. Alternatively, a “metabolic” effect results from changes in the δ 13C of dissolved inorganic carbon (DIC) in the regions adjacent to the site of biomineral formation, and leads to additional changes in skeletal δ 13C. For example, in scleractinian corals symbiotic with zooxanthellae, photosynthesis of symbionts is thought to increase skeletal δ 13C (zooxanthellae mostly fix 12 C), whereas high respiration rates may, in turn, decrease skeletal δ 13C. Plotting skeletal δ 13C and δ 18O for the stalk of M. rotundus (Fig. 4a; our specimen receiving the most attention) revealed a very weak correlation (Fig. 7), i.e. δ 13C changes are not associated with δ 18O changes. This δ 13C negative deviation is most likely due to a “metabolic” effect caused by respiration (McConnaughey, 1989a,b). Indeed, according to Weber and Raup (1966) and Weber (1968) the variations in δ 13C within individuals and across the different echinoderm taxa are independent of diet. These authors suggest that the variations may result from the incorporation of isotopically distinct CO32 − from different sources (most probably from mixing metabolic CO2 and seawater bicarbonate in varying proportions). Weber (1968) suggested that echinoderms fractionate light stable isotopes different than other animal phyla because they lack an efficient, well-developed respiratory system. A specialized respiratory system could reduce isotope exchange in the coelomic perivisceral fluid by adding respiratory carbon dioxide (enriched in 12C and 16O) to seawater bicarbonate (relatively enriched in 13C and 18O), which is the source of the carbonate used for skeletal formation. Crinoids, however, do not have typical respiratory organs. Respiration in these echinoderms may be difficult because of their small size in relation to the large surface area (Hyman, 1955). Respiration takes place from the tube feet by diffusion of oxygen through the body wall. Oxygen and CO2 are transported in internal organs through the perivisceral coelomic fluid (Ausich et al., 1999). The ionic composition of this fluid is similar to that of seawater. However, its pH is much lower, because it contains respiratory carbon dioxide. The answer to the question of why the isotopic composition of cyrtocrinids is similar to that of modern marine cements remains obscure. It is interesting to note, however, that Weber (1968) observed that the mean δ 13C and δ 18O values of echinoid spines, the tests of irregular echinoids, and ophiuroids (Ophiuroidea), are all similar to those for “inorganic calcium carbonate”. He attributed this similarity to a highly efficient respiratory system as well as to the fact that their skeletal calcite constitutes a fairly large proportion of the total body weight and much less respiratory CO2 is produced for a particular unit of the skeleton. Weber (1968) stressed that isotopic composition of the echinoderm skeleton depends on the extent of isotope fractionation between respiratory CO2 and seawater bicarbonate, as well as on the relative quantities of these two components and their isotopic composition. We suggest that a similar mechanism may account for the different isotopic compositions of cyrtocrinid skeletons. In contrast to isocrinids and comatulids, cyrtocrinids have extremely simplified body plan:

Fig. 7. Relationship between δ13C and δ18O values of the stalk of Metacrinus rotundus (specimen no. 3, see Fig. 4a).

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

they lack chambered organs, a glandular axial organ and an aboral nerve system (Heinzeller and Fechter, 1995). Furthermore, these crinoids have reduced or even obliterated arms (Hess, 1999a). Thus, their respiration process, which occurs on the surface of the tube feet, which are located on arms, is probably different. This is consistent with data on metabolic rates of crinoids (Baumiller and LaBarbera, 1989; Hughes et al., 2011). These results show that cyrtocrinids have much lower oxygen consumption rates (0.5 mmol O2 h− 1 gWM − 1) than the representatives of other crinoid orders i.e., Isocrinida (ca. 1 mmol O2 h− 1 gWM− 1), Comatulida (ca. 1–3 mmol O2 h− 1 gWM− 1), Bourgueticrinida (ca. 2 mmol O2 h− 1 gWM− 1). 4.4.1. Possible applications of isotopic compositions to phylogenic problems Our results clearly show that carbon isotope fractionation by crinoids is genetically controlled. Therefore, it seems that isotopes may help resolve some phylogenetic problems. To illustrate this, the following specific example is discussed here. The systematic position of the genus Caledonicrinus is uncertain and remains a matter of debate. Bourseau et al. (1991) classified this genus within Bathycrinidae (Millericrinida). On the other hand, Mironov (2000) included this genus into a new subfamily Caledonicrininae (Bourgueticrinina). Recently, Mironov (2008) raised this subfamily to family status. In the new edition of the Treatise on Invertebrate Paleontology, Pt. 3, Crinoids (Hess and Messing, 2011), this genus is classified into suborder Bourgueticrinina of the order Comatulida. In contrast, a recent molecular analysis provided by Cohen et al. (2004) placed Caledonicrinus amongst cyrtocrinids (Cyrtocrinida). However, a cladistic analysis by the latter authors did not link this genus with any clade. The average δ13C for Caledonicrinus ~+1.9‰ (and ~−1.5‰ in Baumiller, 2001) sharply differs from the isotopic compositions of comatulids, bourgueticrinids and isocrinids (Fig. 5). We therefore suggest that Caledonicrinus should be classified within Cyrtocrinida, which is consistent with the molecular data (Cohen et al., 2004). 4.4.2. Temperature effect The present analysis shows that although crinoids exhibit a large vital effect in the fractionation of carbon isotopes (except the cyrtocrinids), they generally fractionate oxygen isotopes to a lesser degree. Our data suggest a moderate degree of negative correlation between δ18O and mean annual temperature of crinoid habitats. In contrast, the earlier study of Baumiller (2001) suggests a strong negative correlation. We believe that our results are more accurate because they are based on more data (own and from the literature). Our δ 18O-temperature correlation does not allow paleo-temperatures to be reconstructed precisely. However, in Surga Bay, δ 18O changes in the stalk of the shallowest of any living stalked crinoid M. rotundus are similar to the pattern of annual change of the water temperature (Oji, 1989). Based on δ 18O values in all nodal columnals, the latter author estimated that stalks of this species grow at a rate of 30–60 cm y− 1 (details in Oji, 1989). The annual growth rate of our specimen, calculated by comparing δ 18O values in a series of nodals, is only slightly slower at ca. 23 cm y − 1 (Fig.4a). This implies that oxygen isotopes can be used as proxy recording annual growth rates of the stalk of the shallow-sea fossil crinoids that are well preserved. 4.4.3. Implications for applying stable isotopes to evaluate the diagenetic alteration of fossil crinoids Isotopic data for extant crinoids have shown that compositional variation across various crinoid taxa is extremely high. Based on literature data, the maximum δ 13C range is −9 to + 2‰, and the δ 18O range is − 6 to + 3‰. These results allow re-evaluation of the previous hypothesis by Hasiuk and Lohmann (2008) that isotopes may prove useful in characterizing the degree of post-depositional diagenesis of fossil crinoids. Hasiuk and Lohmann (2008) analyzed a suite of Mississippian crinoids and observed that their isotopic composition does not overlap with extant crinoids. They concluded that these fossil crinoids were altered by diagenesis. However, the Paleozoic

139

crinoids differ morphologically from their post-Paleozoic descendants, and they may have fractionated light stable isotopes differently. Interestingly, the extant cyrtocrinids exhibit similar isotope values to some of these Mississippian crinoids. If the Mississippian samples are not biased by diagenesis, this may support the hypothesis (based on the conclusion that major physiological differences between some crinoid clades affect 13C fractionation) that cyrtocrinoids diverged from the crinoid stem group much earlier than had been thought (i.e. in the Late Paleozoic). Indeed, Permian crinoid fauna from Timor (such as Prophyllocrinus, Proapsidocrinus, Paleoholopus) shows strong similarities to the cyrtocrinids from the Jurassic and Lower Cretaceous of Europe (Hess, 1999b). Recent molecular data of cyrtocrinids are uncertain and indicate that their position is either relatively basal, or with isocrinids or hyocrinids (Rouse et al., 2006; Rouse personal communication, 2009). It should be stressed, however, that the isotopic analysis of the fossil echinoderms is extremely difficult. Their porous skeletons tend to be filled by cement which may have an isotopic signature that differs from that of echinoderm skeletons. Using a dental drill to extract powder from 0.5 mm 2 areas of fossil crinoids by Hasiuk and Lohmann (2008) might be insufficient to address this problem.

5. Conclusions The stable carbon (δ 13C) and oxygen (δ 18O) isotope composition of the skeletons of extant stalked crinoids is highly variable at different scale lengths. Within-ossicle variations may be of similar magnitude to that observed within individuals and across various taxa of the same order. Our specimens were collected from the deep-sea environments, where there is almost no significant variability in temperature and salinity. Nevertheless, we observed variations in δ 13C and δ 18O within a single ossicle as well as within the skeleton from a single individual. We also observe significant compositional differences between homologous skeletal parts of different isocrinid taxa from the same environment. These observations suggest that biological factors play an important role in controlling the distribution of light stable isotopes. The differences in δ 13C between two major crinoid clades may amount to ~ 10‰, reflecting distinct physiological differences between those groups. Despite this biological control, crinoids still record in their skeletons an environmental signal, i.e. a temperature effect is detectable in the δ 18O of their skeleton.

Acknowledgments We would like to thank: T. K. Baumiller (Department of Geological Sciences, University of Michigan), T. Oji (The Nagoya University Museum), N. Améziane (Muséum National d'Histoire Naturelle, Paris), M. Wisshak and K. Haase (Geozentrum Nordbayern, Erlangen, Germany) for providing the crinoid specimens. We wish to also thank: A. N. Mironov (P.P. Shirshov Institute of Oceanology, Russian Academy of Sciences, Moscov) for providing some literature, G.W. Rouse (Scripps Institution of Oceanography, UCSD) for providing unpublished data on mitochondrial and nuclear gene sequences of crinoids and H. Hess (Naturhistorisches Museum, Basel) for providing unpublished data on revised Treatise. M. Bender (Princeton University) comments and suggestions improved the clarity of the presentation. We also thank two anonymous referees for providing us with constructive reviews. This work was supported in part by ERC-Advanced grant 246749 (Biocarb).

Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.chemgeo.2011.10.014.

140

P. Gorzelak et al. / Chemical Geology 291 (2012) 132–140

References Ausich, W.I., Brett, C.E., Hess, H., Simms, M.J., 1999. Crinoid form and function. In: Hess, H., Ausich, W.I., Brett, C.E., Simms, M.J. (Eds.), Fossil Crinoids. Cambridge University Press, Cambridge, pp. 3–31. Baumiller, T.K., 2001. Light stable isotope geochemistry of the crinoid skeleton and its use in biology and paleobiology. In: Barker, M. (Ed.), Echinoderms 2000. Balkema, Lisse, pp. 107–112. Baumiller, T.K., Labarbera, M., 1989. Metabolic rates of Caribbean crinoids (Echinodermata), with special reference to deep-water stalked and stalkless taxa. Comparative Biochemistry and Physiology 93A, 391–394. Beier, C. 2006. The magmatic evolution of oceanic plateaus: a case study from the Azores. Ph.D. thesis, University of Kiel, Germany. pp. 1–128. Bourseau, J.-P., Améziane-Cominardi, N., Avocat, R., Roux, M., 1991. Echinodermata: Les Crinoïdes pédonculés de Nouvelle-Calédonie. In: Crosnier, A. (Ed.), Résultats des Campagnes MUSORSTOM, 8: Mém. Mus. natn. Hist. nat. Paris. (A), 151, pp. 229–333. 11 text-fig., 16 pl. Brower, J.C., 1974. Ontogeny of camerate crinoids. Paleontological Contributions Paper, 72. University of Kansas, pp. 1–53. Cohen, B.L., Ameziane, N., Eleaume, M., De Forges, B.R., 2004. Crinoid phylogeny: a preliminary analysis (Echinodermata: Crinoidea). Marine Biology 144 (3), 605–617. David, J. 1998. Adaptation morphologique, croissance et production bioclastique chez les crinoďdes pédonculés actuels et fossiles (Pentacrines et Millericrinina). Application paléoécologique aux gisements du Jurassique supérieur des Charentes et du nordest du bassin de Paris. Travaux Universitaires. PhD thesis, Université de Reims, Reims, France (Université de Soutenance). pp. 1–551. Goodfriend, G.A., Margaritz, M., 1987. Carbon and oxygen isotope composition of shell carbonate of desert land snails. Earth and Planetary Science Letters 86, 377–388. Gonzalez, A., Lohmann, K.C., 1985. Carbon and oxygen isotopic composition of Holocene reef carbonates. Geology 13, 811–814. Hasiuk, F.J., Lohmann, K.C., 2008. Mississippian paleocean chemistry from biotic and abiotic carbonate, Muleshoe Mound, Lake Valley Formation, New Mexico, USA. Journal of Sedimentary Research 78, 147–160. Heinzeller, T., Fechter, H., 1995. Microscopical anatomy of the Cyrtocrinid Cyathidium meteorensis (sive foresti) (Echinodermata, Crinoidea). Acta Zoologica Stockholm 76, 25–34. Hess, H., 1999a. Tertiary. In: Hess, H., Ausich, W.I., Brett, C.E., Simms, M.J. (Eds.), Fossil Crinoids. Cambridge University Press, Cambridge, pp. 233–244. Hess, H., 1999b. Permian. In: Hess, H., Ausich, W.I., Brett, C.E., Simms, M.J. (Eds.), Fossil Crinoids. Cambridge University Press, Cambridge, pp. 160–163. Hess, H., Messing, Ch.G., 2011. In: Selden, P.A. (Ed.), Treatise on Invertebrate Paleontology, Part T, Echinodermata 2 Revised, Crinoidea, vol. 3. The University of Kansas Paleontological Institute, Lawrence Kansas. xxix + 261 p., 112 fig. Hughes, S.J.M., Ruhl, H.A., Hawkins, L.E., Hauton, C., Boorman, B., Billett, D.S.M., 2011. Deep-sea echinoderm oxygen consumption rates and an interclass comparison of metabolic rates in Asteroidea, Crinoidea, Echinoidea, Holothuroidea and Ophiuroidea. Journal of Experimental Biology 214, 2512–2521. Hyman, L.H., 1955. The invertebrates: Echinodermata IV. McGraw-Hill, pp. 1–763. Kitazawa, K., Oji, T., Sunamura, M., 2007. Food composition of crinoids (Crinoidea: Echinodermata) in relation to stalk length and fan density: their paleoecological implications. Marine Biology 152, 959–968. Leaman, K.D., Molinari, R.L., Vertes, P.S., 1987. Structure and variability of the Florida Current at 27°N: April 1982–July 1984. Journal of Physical Oceanography 17, 565–583. McConnaughey, T., 1989a. 13C and 18O isotopic disequilibrium in biological carbonates: I. Patterns. Geochimica et Cosmochimica Acta 53, 151–162.

McConnaughey, T., 1989b. 13C and 18O isotopic disequilibrium in biological carbonates: II in vitro simulation of kinetic isotope effects. Geochimica et Cosmochimica Acta 53, 163–171. McConnaughey, T., Gillikin, D.P., 2008. Carbon isotopes in mollusk shell carbonates. Geo-Marine Letters 28 (5–6), 287–299. Messing, C.G., Roux, M., Améziane, N., Baumiller, T.K., 2007. In situ stalk growth rates in tropical western Atlantic sea lilies (Echinodermata: Crinoidea). Journal of Experimental Marine Biology and Ecology 353, 211–220. Mironov, A.N., 2000. New taxa of stalked crinoids from the suborder Bourgueticrinina (Echinodermata, Crinoidea). Zoologichesky Zhurnal 79, 712–728 (in Russian). Mironov, A.N., 2008. Stalked crinoids of the family Bathycrinidae (Echinodermata) from the eastern Pacific. Invertebrate Zoology 5 (2), 133–153. Ohta, S., 1983. Photographic census of large-sized benthonic organisms in the bathyal zone of Suruga Bay, central Japan. Bulletin of the Ocean Research Institute, University of Tokyo 15, 1–244. Oji, T., 1989. Growth rate of stalk of Metacrinus rotundus (Echinodermata: Crinoidea) and its functional significance. Journal of the Faculty of Science University of Tokyo 22, 39–51. Rasmussen, H.W., 1978. Articulata. In: Moore, R.C., Teichert, C. (Eds.), Treatise on Invertebrate Paleontology. : Pt. T, Echinodermata 2, vol. 3. Geological Society of America and University of Kansas Press, pp. T813–T928. Richter, D.K., Bruckschen, P., 1998. Geochemistry of recent tests of Echinocyamus pusillus: constraints for temperature and salinity. Carbonates Evaporites 13 (2), 157–167. Rouse, G.W., Jermin, L.S., Messing, C.G., 2006. Phylogeny of extant Crinoidea based on mitochondrial and nuclear gene sequences. Abstract, 12th International Echinoderm Conference, August 7–11, 2006. Durham NH, pp. 54–55. Roux, M., Renard, M., Améziane, N., Emmanuel, L., 1995. Zoobathymetrie et composition chimique de la calcite des ossicules du pedoncule des crinoides. Comptes Rendus de l'Académie des Sciences Paris Série IIa 321, 675–680. Simms, M.J., 1989. Columnal ontogeny in articulate crinoids and its implications for their phylogeny. Lethaia 22, 61–68. Sullivan, C.H., Krueger, H.W., 1981. Carbon isotope analysis of separate chemical phases in modem and fossil bone. Nature 292, 333–335. Takahashi, S., Tanabe, S., Kubodera, T., 1997. Butyltin residues in deep-sea organisms collected from Suruga Bay, Japan. Environmental Science and Technology 31, 3103–3109. Weber, J.N., 1968. Fractionation of the stable isotopes of carbon and oxygen in marine calcareous organisms—the Asteroidea, Ophiuroide and Crinoidea. Geochimica et Cosmochimica Acta 32, 33–70. Weber, J.N., Raup, D.M., 1966. Fractionation of the stable isotopes of carbon and oxygen in marine calcareous organisms—the Echinoidea. Part 1. Variation of C13 and O18 content within individuals. Geochimica et Cosmochimica Acta 30, 681–703. Weber, J.N., Raup, D.M., 1968. Composition of C13/C12 and O18/O16 in the skeletal of recent and fossil echinoids. Journal of Paleontology 42, 37–50. Wisshak, M., Neumann, C., Jakobsen, J., Freiwald, A., 2009. The ‘living-fossil community’ of the cyrtocrinid Cyathidium foresti and the deep-sea oyster Neopycnodonte zibrowii (Azores Archipelago). Palaeogeography, Palaeoclimatology, Palaeoecology 271, 77–83. Vail, L., 1989. Arm growth and regeneration in Oligometra serripinna (Echinodermata: Crinoidea) at Lizard Island, Great. Barrier Reef. Journal of Experimental Marine Biology and Ecology 130, 189–204.