Quaternary Geochronology 5 (2010) 519–532 Contents lists available at ScienceDirect Quaternary Geochronology journal homepage: www.elsevier.com/locate/quageo Research Paper Radiocarbon dating of small terrestrial gastropod shells in North America Jeffrey S. Pigati a, *, Jason A. Rech b, Jeffrey C. Nekola c a U.S. Geological Survey, Denver Federal Center, Box 25046, MS-980, Denver CO 80225, USA Department of Geology, Miami University, Oxford, OH 45056, USA c Department of Biology, University of New Mexico, Albuquerque, NM 87131, USA b a r t i c l e i n f o a b s t r a c t Article history: Received 26 May 2009 Received in revised form 20 January 2010 Accepted 21 January 2010 Available online 29 January 2010 Fossil shells of small terrestrial gastropods are commonly preserved in wetland, alluvial, loess, and glacial deposits, as well as in sediments at many archeological sites. These shells are composed largely of aragonite (CaCO3) and potentially could be used for radiocarbon dating, but they must meet two criteria before their 14C ages can be considered to be reliable: (1) when gastropods are alive, the 14C activity of their shells must be in equilibrium with the 14C activity of the atmosphere, and (2) after burial, their shells must behave as closed systems with respect to carbon. To evaluate the first criterion, we conducted a comprehensive examination of the 14C content of the most common small terrestrial gastropods in North America, including 247 AMS measurements of modern shell material (3749 individual shells) from 46 different species. The modern gastropods that we analyzed were all collected from habitats on carbonate terrain and, therefore, the data presented here represent worst-case scenarios. In sum, w78% of the shell aliquots that we analyzed did not contain dead carbon from limestone or other carbonate rocks even though it was readily available at all sites, 12% of the aliquots contained between 5 and 10% dead carbon, and a few (3% of the total) contained more than 10%. These results are significantly lower than the 20–30% dead carbon that has been reported previously for larger taxa living in carbonate terrain. For the second criterion, we report a case study from the American Midwest in which we analyzed fossil shells of small terrestrial gastropods (7 taxa; 18 AMS measurements; 173 individual shells) recovered from late-Pleistocene sediments. The fossil shells yielded 14C ages that were statistically indistinguishable from 14C ages of well-preserved plant macrofossils from the same stratum. Although just one site, these results suggest that small terrestrial gastropod shells may behave as closed systems with respect to carbon over geologic timescales. More work on this subject is needed, but if our case study site is representative of other sites, then fossil shells of some small terrestrial gastropods, including at least five common genera, Catinella, Columella, Discus, Gastrocopta, and Succinea, should yield reliable 14C ages, regardless of the local geologic substrate. Published by Elsevier B.V. Keywords: Radiocarbon Land snails Limestone effect Chronology Quaternary 1. Introduction Gastropods are one of the most successful animal groups on Earth, with at least 70,000 extant species occupying terrestrial, marine, and freshwater habitats. Globally, terrestrial gastropods encompass at least 35,000 species (Barker, 2001), span 4–5 orders of magnitude in shell volume, and represent a variety of trophic levels, including polyphagous detritivores, herbivores, omnivores, and carnivores (Kerney and Cameron, 1979; Burch and Pearce, 1990). They are so exceptionally diverse in their appearance, ecology, and physiology that determining their phylogenetic relationships from conchological and/or anatomical characteristics * Corresponding author. Tel.: þ1 303 236 7870; fax: þ1 303 236 5349. E-mail address: jpigati@usgs.gov (J.S. Pigati). 1871-1014/$ – see front matter Published by Elsevier B.V. doi:10.1016/j.quageo.2010.01.001 remains difficult and controversial (e.g., Ponder and Lindberg, 1997 and references therein). It is clear, however, that the preference for terrestrial habitats of North American gastropods developed independently in three of six basal clades (Neritomorpha, Caenogastropoda, and Heterobranchia), with the informal group Pulmonata representing more than 99% of the continental fauna. Of the Pulmonata, the most common size class1 in both modern and fossil assemblages are individuals with adult shells that are <10 mm in maximum dimension (Nekola, 2005) (Fig. 1). 1 Size classes of gastropods are categorized by the maximum shell dimension (length or diameter) as follows: large (>20 mm), medium (10–20 mm), small (5– 10 mm), minute (2–5 mm), and micro (<2 mm). Although the size of the gastropods targeted in this study range from small to micro, for simplicity, we refer to them collectively as ‘‘small’’. 520 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 Fig. 1. Photographs of select small terrestrial gastropods included in this study (1 mm bar in each panel for scale). (a) Cochlicopa lubricella, (b) Columella columella alticola, (c) Discus macclintockii, (d) Euconulus fulvus, (e) Hawaiia miniscula, (f) Hendersonia occulta, (g) Punctum minutissimum, (h) Pupilla muscorum, (i) Strobilops labyrinthica, (j) Vallonia gracilicosta, (k) Vertigo elatior, and (l) Zonitoides arboreus. Today, small terrestrial gastropods occupy and thrive in incredibly diverse habitats, from marshes, wet meadows, and grasslands to upland forests and tundra. Species are known from all continents, save Antarctica, and occupy almost all climate regimes except hyperarid deserts and the high Arctic. Their distribution in the fossil record is equally diverse. Gastropod shells are commonly preserved in wetland, alluvial, loess, and glacial deposits, as well as within sediments at archeological sites worldwide (e.g., Evans, 1972). But even though their distribution is widespread and their aragonitic shells contain w12% by weight carbon, terrestrial gastropods are often avoided for 14C dating because many taxa incorporate 14C-deficient (or ‘‘dead’’) carbon from limestone and other carbonate rocks when building their shells. This phenomenon, referred to as the ‘‘Limestone Problem’’ by Goodfriend and Stipp (1983), can cause 14C ages of gastropod shells to be as much as w3000 yrs too old. Despite the Limestone Problem, geochronologists have continued to investigate the possibility of using terrestrial gastropod shells for 14 C dating because of their widespread occurrence and potential for dating Quaternary sediments. Most 14C studies of gastropod shells have found that gastropods consistently incorporate dead carbon from limestone in their shells when it is available (Frye and Willman, 1960; Leighton, 1960; Rubin et al., 1963; Tamers, 1970; Evin et al., 1980; Goodfriend and Hood, 1983; Goodfriend and Stipp, 1983; Goslar and Pazdur, 1985; Yates, 1986; Goodfriend, 1987; Zhou et al., 1999; Quarta et al., 2007; Romaniello et al., 2008). These studies, however, were generally limited to a few individual gastropods collected from a small number of sites, and were biased toward large taxa and warm climates. Brennan and Quade (1997) analyzed a number of small terrestrial gastropod taxa and found that small shells generally yielded reliable 14C ages for late-Pleistocene paleowetland deposits in the American Southwest. Pigati et al. (2004) followed by measuring the 14C activities of a suite of small gastropods living in alluvium dominated by Paleozoic carbonate rocks in Arizona and Nevada and found that while some of the small gastropods did incorporate dead carbon from limestone when building their shells, others did not. Based in part on these initial positive results, small terrestrial gastropod shells have been used recently to date Quaternary wetland and lacustrine deposits in the Americas (e.g., Pedone and Rivera, 2003; Placzek et al., 2006; Pigati et al., 2009). However, it is unclear if the results obtained from modern gastropods collected from a limited number of sites in the American Southwest can be extrapolated to all geologic, ecologic, and climatic environments. Moreover, it is not known if results for one taxonomic level (family, genus, or species) can be extrapolated to other members of the 521 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 same level living elsewhere, or even between individuals living within the same population. Here we report the results of a comprehensive analysis of the Limestone Problem for small terrestrial gastropods from 163 localities in North America (Fig. 2). All samples that we analyzed were collected from habitats on carbonate terrain and, therefore, the data reported here represent worst-case scenarios. In addition, we measured the 14C activities of a number of fossil shells recovered from well-dated sediments at a late-Pleistocene site in the American Midwest as a case study to determine if the shells remain closed systems with respect to carbon over geologic timescales. Positive results for both tests for a particular taxon would allow us to consider 14C ages derived from fossil shells of that taxon to be reliable, regardless of the local geologic substrate. 2. Shell carbonate and 14 C dating All materials (organic and inorganic) that yield reliable 14C ages have two common characteristics. First, the initial 14C activity of the material – a plant, for example – was in equilibrium with atmospheric 14C at the time that it was alive. In other words, the 14C activity of a plant that lived T yrs ago was the same as the 14C activity of the atmosphere T yrs ago (after accounting for isotopic fractionation). Second, after death, the material behaved as a closed system; carbon was neither added to nor removed from the sample material. If both of these criteria are met, then the measured 14C activity is a function of only two parameters: the initial 14C activity of the atmosphere and the amount of time elapsed since the death of the organism. The measured 14C activity and the 14C age of the material are related by the familiar decay equation A ¼ Ao elt (1) 14 where A and Ao are the measured and initial C activities of the material, respectively, l is the decay constant, and t is the time elapsed since the death of the organism. Conventional radiocarbon ages assume that the atmospheric 14C activity is invariant through time (i.e., Ao ¼ 1). Radiocarbon ages can be converted to calendar year ages to account for temporal variations in the 14C activity of the atmosphere (Reimer et al., 2009). Fig. 2. Locations of modern localities (red dots) and the fossil locality at the Oxford East outcrops in southwestern Ohio (star). Modern localities and collection information are listed in Tables S1 and S2, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article). 522 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 or granules which then dissolve in their stomach acid to produce CO2. As before, the dead carbon from the rocks is introduced to the hemolymph, passed on to the extrapallial fluid, and ultimately incorporated in the shell carbonate. Dead carbon from limestone can account for up to w30% of the total carbon in shells of large terrestrial gastropods. (Goodfriend and Stipp, 1983). 2 Deviation from true 14C age (ka) 1 0 -1 -2 14 C ages of shell carbonate -3 2.2. Effects of carbon sources on -4 In most environments, the 14C activities of live plants are in equilibrium with atmospheric carbon. Gastropods that obtain their shell carbon from live plants and the air, therefore, should yield reliable 14C ages, assuming they behave as closed systems after burial (Fig. 3). Gastropods that consume organic detritus (i.e., decaying plant litter) typically do not pose a significant problem for 14 C dating because the time between plant death, its incorporation into decomposition products, and consumption by gastropods is usually quite short, on the order of a few yrs. The 14C activity of water that is available for consumption by terrestrial gastropods (e.g., dew, standing water, precipitation) is at or near equilibrium with atmospheric 14C and, therefore, water is unlikely to introduce a significant error to 14C ages of terrestrial gastropod shells. Exceptions include gastropods living directly adjacent to springs that discharge waters from deeply-circulating carbonate aquifers, lakes or rivers with significant hard water effects, or in active volcanic areas where 14C-deficient CO2 in surface waters may be abundant (e.g., Riggs, 1984; Grosjean, 1994). 14 C ages of gastropods living in such areas should be evaluated carefully. The incorporation of 14C from limestone and other carbonate rocks can present a significant problem for 14C dating of terrestrial gastropod shells. The 14C activity of atmospheric carbon, plants, and water consumed by gastropods is essentially the same, w100% modern carbon (pMC). In contrast, because most carbonate rocks are of pre-Quaternary age, their 14C activity is typically 0 pMC. Thus, for 14C dating, the magnitude of the potential error introduced by carbonate rocks is a direct function of the amount of carbon from rocks that is incorporated in the gastropod shell (Fig. 3). Unfortunately, a simple correction that accounts for the incorporation of 14 C-deficient carbon in gastropod shells is not possible because we cannot know a priori how much of the shell carbon was derived from carbonate rocks versus other sources. Thus, to be confident in 14 C ages derived from terrestrial gastropod shells, it is imperative that we identify and avoid taxa that incorporate dead carbon from rocks altogether. -5 -6 -7 -8 0 5 10 15 20 25 True 14C age (ka) 30 35 40 Fig. 3. Modeled deviation from the true 14C age for four scenarios: (1) closed-system behavior and no dead carbon (thick solid line), (2) closed-system behavior and 10% dead carbon (dashed line), (3) open-system behavior equivalent to 1% modern carbon contamination and 10% dead carbon (thin solid line), and (4) open-system behavior equivalent to 1% modern carbon contamination and no dead carbon (dotted line). 2.1. Sources of shell carbon In order to evaluate the validity of a 14C age of a given sample, the 14C contents of the original sources of the carbon and their contribution to the total carbon content must be known. Carbon in gastropod shell carbonate originates from as many as four different sources: atmospheric CO2, food, water, and carbonate rocks. Gastropods incorporate atmospheric CO2 in their shell carbonate via respiration. Respired CO2 is introduced to the bicarbonate pool in the gastropod’s hemolymph and passed along to the extrapallial fluid, from which the shell carbonate is ultimately precipitated (Wilbur, 1972). Estimates of the contribution of atmospheric CO2 to gastropod shell carbonate vary between negligible (Stott, 2002), 16–48% (Romaniello et al., 2008), and 30– 60% (Goodfriend and Hood, 1983). Carbon from food sources (e.g., living plants, fungi, organic detritus) is incorporated into the extrapallial fluid through two mechanisms: direct digestion and breakdown of urea. When gastropods consume and digest food, carbon is introduced to the hemolymph and passed along to the extrapallial fluid in the same manner as atmospheric CO2. There it mixes with atmospheric carbon before becoming incorporated in the shell carbonate (Wilbur, 1972). Carbon derived from urea takes a more indirect pathway. Urea that is not expelled by the gastropod breaks down into CO2 and NH3 via a urease reaction (Stott, 2002). The resulting CO2 is then reintroduced directly to the extrapallial fluid and ultimately incorporated into the shell carbonate. Estimates of the amount of carbon derived from plants, either directly or indirectly through urea, vary between 25 and 40% (Goodfriend and Hood, 1983), 36–73% (Romaniello et al., 2008), and w100% (Stott, 2002). Terrestrial gastropods ingest water from multiple sources, including dew, soil moisture, standing water, and precipitation, all of which contain some amount of dissolved inorganic carbon (DIC). Water is taken up through the foot of the gastropod by contact rehydration (Balakrishnan and Yapp, 2004) and introduced to the hemolymph before being passed on to the extrapallial fluid. Pigati et al. (2004) found that aqueous carbon sources account for w10% of the shell carbon for one species of Catinella, but it is not known if this value is constant across the entire Succineidae family. To our knowledge, data for other terrestrial taxa do not exist. Finally, some terrestrial gastropods are able to scrape carbonate rocks (limestone, dolomite, soil carbonate), and ingest the powder 2.3. Open-system behavior Even if some terrestrial gastropods consistently manage to avoid the Limestone Problem regardless of the local geologic substrate or environmental conditions, there is another hurdle that must be overcome before we can confidently use their shells for 14C dating. That is, gastropod shells must remain closed systems with respect to carbon after burial. For reliable 14C dating, the pool of carbon atoms measured during the 14C dating process must consist solely of carbon atoms that originally resided in the shell. Thus, following burial, shells must resist the addition or exchange of 14C atoms with the local environment. Shells that exhibit open-system behavior typically yield 14C ages that are too young, and the magnitude of the error depends upon the degree of such behavior (Fig. 3). Previous work has suggested that 14C ages from small terrestrial gastropod shells recovered from fossil deposits in arid environments may be reliable back to at least w30,000 14C yrs B.P., but a small degree of open-system behavior appears to compromise 14C dates obtained from shells older than this (Pigati et al., 2009). In J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 Fig. 4. Photographs of the three genera of the Succineidae family: Catinella (left panels), Oxyloma (middle panels), and Succinea (right panels); all are w10 mm in length. The simple shells of the three genera contain few diagnostic characteristics and, therefore, species-level identification is based on soft-body reproductive organ morphology, which is rarely preserved in the fossil record. this study, we analyzed fossil shells from the American Midwest because of their abundance in Quaternary deposits in the region, the presence of multiple calcareous substrates (Paleozoic limestone, calcareous till, and loess), and the humid climate (annual precipitation in southwestern Ohio is w100 cm yr1). If fossil shells exhibit even a small degree of open-system behavior in arid environments, it may be exacerbated and, therefore, more easily detected in humid environments where interaction between shells and DIC in ground water is more prevalent. 3. Methods 3.1. Live gastropods Previous ecological sampling by one of us (JCN) has resulted in an extensive collection of modern terrestrial gastropods from North America, constituting w250 taxa and over 470,000 individuals from more than 1000 modern environments. Gastropods were collected at each site from a representative 100–1000 m2 area by hand collection of larger taxa and litter sampling of smaller taxa, which provides the most complete assessment of site faunas (Oggier et al., 1998; Cameron and Pokryszko, 2005). Collections were made at places of high mollusc density, such as loosely compacted leaf litter lying on top of highly compacted damp soil or humus (Emberton et al., 1996). Litter was removed by hand and sieved by shaking, tapping, or other agitation in the field using a shallow sieve (ASTME #10; 2.0 mm mesh) nested inside a second sieve (ASTME #30; 0.6 mm mesh). The process was continued for 15–60 min during which time 50–500 mg of material was collected and retained. Gastropods and detritus were dried at room temperature in the laboratory and then hand-picked against a neutral background. All shells, shell fragments, and slug plates were removed and identifiable material was assigned to species using JCN’s reference collection. Nomenclature generally follows that of Hubricht (1985) with updates and corrections by Nekola (2004). 523 We selected 247 aliquots of shell material (3749 individual shells) from 163 sites across the United States and southern Canada for 14C analysis. Nearly all of the specimens that we chose for analysis were collected live, but at a few sites, only recently-dead gastropods were available, which were identified by a translucent appearance or the retention of color in the shells. Shells of small gastropods that are dead for more than a year or so become increasingly white and opaque with time (J. Nekola, unpublished data), and were excluded from our study. In the fossil record, species-level identification of fossil shells is possible for most small terrestrial gastropods and, therefore, the results of our investigation of modern gastropods can be applied directly to the fossil record. An exception is the Succineidae family, which is composed of three genera (Catinella, Oxyloma, and Succinea) that are difficult to differentiate in modern faunas, let alone the geologic record (Fig. 4). Their simple shells exhibit few diagnostic characteristics and, therefore, species-level identification is based on soft-body reproductive organ morphology, which is rarely preserved in the fossil record. This presents a significant problem for geochronologists; that is, can we be confident in 14C ages derived from shells from any taxon within the Succineidae family, or do we need to target a specific genus or species? To address this issue, we measured the 14C activity of 100 aliquots of gastropod shell material (802 individual shells) for twelve species of the Succineidae family to determine the level of identification (i.e., family, genus, or species) required to apply our results to the fossil record. We prepared aliquots of modern shell material for 14C analysis at the University of Arizona Desert Laboratory (JSP) and Miami University (JAR). We selected multiple shells at random for X-ray diffraction (XRD) analysis using a Siemens Model D-500 diffractometer to verify that only shell aragonite remained prior to preparation for 14C analysis. There was no evidence of primary or secondary calcite in any of the shells that we analyzed. When possible, shells were broken, the adhering soft parts were removed using forceps, and the shells were treated with 6% NaOCl for 18– 24 h at room temperature to remove all remnants of organic matter. Shells were not powdered during pretreatment to minimize the potential for adsorption of atmospheric 14C (Samos, 1949). We selectively dissolved some of the shells by briefly introducing dilute HCl to remove secondary carbonate (dust) from primary shell material. Shells were washed repeatedly in ASTM Type 1, 18.2 MU (hereafter ‘‘ultrapure’’) water, sonicated for a few seconds to remove adhered solution, washed again with ultrapure water, and dried in a vacuum oven overnight at w70  C. Shell aragonite was converted to CO2 using 100% H3PO4 under vacuum at either 50 or 75  C until the reaction was visibly complete (w1 h). Water, SOx, NOx, and halide species were removed using passive Cu and Ag traps held at w600  C and the resulting CO2 was split into two aliquots. One aliquot was converted to graphite by catalytic reduction of CO (modified after Slota et al., 1987) and submitted to the Arizona-NSF Accelerator Mass Spectrometry (AMS) facility for 14C analysis. The second aliquot was submitted for d13C analysis in order to correct the measured 14C activity of the shell carbonate for isotopic fractionation. 14 C data for modern gastropods are presented as D14C values in per mil (Stuiver and Polach, 1977; Reimer et al., 2004) and analytical uncertainties are reported at the 2s (95%) confidence level. d13C values are given in the usual delta (d) notation as the per mil deviation from the VPDB standard. Analytical uncertainties for d13C measurements are less than 0.1& based on repeated measurements of standards. We also measured the 14C content of several gastropod bodies for comparison to their corresponding shells. The bodies were treated with 10% HCl for 15–30 min to remove any remnants of the carbonate shell, rinsed repeatedly, and dried in a vacuum oven at 524 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 C values of gastropod diets To quantify the amount of carbon in a gastropod shell that was derived from limestone or other carbonate rocks, it is necessary to compare the measured D14C values of shells with D14C values of the gastropod diet. D14C values of live plants consumed by gastropods are identical to D14C values of the atmosphere (Fig. 5a), which were calculated using 14C data averaged over the Northern Hemisphere (Hua, 2004). We assigned a 5& uncertainty to the atmospheric values to account for short-term, regional variations (Hsueh et al., 2007) and small changes in atmospheric 14C values that occur at a given site during the short (usually annual) lifespan of the small gastropods (Manning et al., 1990; Meijer et al., 1995). Plant detritus from previous years’ primary production has a higher 14C activity than live plants because of the 14C ‘‘bomb-spike’’ (Hua, 2004). Simply comparing the D14C values of shells with the D14C of the atmosphere or live plants during the year that the gastropod was alive, therefore, would ignore the potential impact of detritus in the gastropod diet. We estimated the amount of detritus from a given year that was available for consumption using a wide range of carbon turnover rates (0.2–0.002 yr1; Fig. 5b), which are applicable to O horizons and the upper few centimeters of A horizons in which small terrestrial gastropods typically live (Guadinski et al., 2000; Torn et al., 2005; Brovkin et al., 2008). We intentionally chose a wide range of carbon turnover rates, which are equivalent to mean residence times for carbon of 5–500 yrs, to encompass the potentially wide range of detritus ages at the 163 localities included in our study. We then used Monte Carlo simulation to generate 10,000 values for the D14C of the gastropod diets for each of the past 13 yrs (the time span of our collections), which allowed two factors to vary randomly: the carbon turnover rate (0.2–0.002 yr1) and the age of the detritus fraction included in the diet (range of 0–1000 yrs). We also ran simulations in which we let the age of the detritus vary up to 10,000 yrs, but the results did not change significantly. We took the average and standard deviation of the 10,000 generated values as the gastropod diet D14C value for each year of collection between 1996 and 2008 (Fig. 5c). All uncertainties are reported at the 2s (95%) confidence level. As expected, the modeled D14C value of the gastropod diet for a given year is slightly higher than the atmospheric D14C value for the same year. Uncertainties associated with the modeled values are relatively large, on the order of w35%, because of the large range of detritus D14C values that could be present at a given site. Shells with D14C values that are lower than the modeled D14C value of the gastropod diet during the year in which the gastropod was alive indicate the presence of dead carbon from limestone or other carbonate rocks. When applicable, the difference between the D14C values was converted into 14C yrs to estimate the ‘‘limestone effect’’, which represents the potential error introduced by the incorporation of dead carbon in the shells. The magnitude of the limestone effect should be considered a maximum value because the calculation assumes that all of the dead carbon came from carbonate rocks, rather than older (but not infinitely-aged) organic matter. Because of the uncertainties associated with modeling the D14C of the gastropod diet, we are unable to discern limestone effects smaller than w300 14C yrs. ∆ 14C of atmosphere (‰) 14 250 200 150 100 50 0 1980 1985 1990 1995 2000 2005 2010 Year b 1.0 0.9 Fraction of detritus available for consumption 3.2. Modeling of a 0.8 0.7 0.6 0.002 yr-1 0.5 0.2 yr-1 0.4 0.3 0.05 yr-1 0.2 0.02 yr-1 0.1 0.01 yr-1 0.0 0 200 400 600 800 1000 Time after plant death (years) c 250 ∆14C of gastropod diet (‰) w70  C. The dried bodies were placed in 6 mm quartz tubes along with w100 mg of cupric oxide (CuO) and a small piece (1 mm  5 mm) of silver foil, all of which were pre-combusted at 900  C for 4–6 h. The tube was evacuated, sealed with a glassblower’s torch, and combusted offline at 900  C. The resulting CO2 gas was isolated and converted to graphite as above. 200 150 100 50 0 1993 1998 2003 2008 Date of collection Fig. 5. Modeling results for (a) the 14C activity of the gastropod diet using measured atmospheric values for the northern hemisphere (after Hua, 2004), (b) carbon turnover rates (CTRs) ranging from 0.2 to 0.002 yr1, and (c) Monte Carlo simulation to generate estimates of the D14C values of gastropod diets for each of the past 13 yrs. 3.3. Fossil gastropods Fossil gastropod shells were collected from glacial deposits at the Oxford East glacial outcrops in southwestern Ohio as a case study to determine if the shells remain closed systems with respect to carbon over geologic timescales. These outcrops contain a series of glacial diamictons from the Miami Lobe of the Laurentide Ice J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 Sheet that are separated by thin (3–5 cm) units of calcareous organic-rich silt that contain gastropods, plant macrofossils, and rooted tree stumps. AMS 14C dating of plant macrofossils and in situ tree stumps has shown that the age of this unit is between w20,100 and 21,400 14C yrs B.P. (Eckberg et al., 1993; Lowell, 1995). Gastropod taxa identified previously from this unit include Columella columella, Discus cronkhitei, Euconulus fulvus, Hendersonia occulta, Pupilla muscorum, Vertigo elatior, and multiple Succineidae taxa (Dell, 1991). Gastropod-bearing sediment was collected and placed in deionized water with a deflocculant for several days to soften the sediment enough to pass through a 0.5 mm sieve. A few samples were placed in an ultrasonic bath for w1 h to further disaggregate the sediment. Fossil shells were hand-picked from the retained fraction, placed in a beaker of ultrapure water, subjected to an ultrasonic bath for a few seconds, and then repeatedly dunked in a second beaker of ultrapure water to remove sediment that adhered to the shell surface or was lodged within the shell itself. The recovered shells were broken and examined under a dissecting microscope to ensure that the interior whorls were free of secondary carbonate and detritus. Fossil shells that were free of detritus were then processed for 14C in the same manner as the modern specimens, including random selection of shells for XRD analysis. None of the fossil shells that we analyzed contained measurable quantities of either primary or secondary calcite. Organic samples, which included bark, charcoal, plant fragments, and wood, were subjected to a standard acid–base–acid (ABA) chemical pretreatment with 1N HCl (1 h at 60  C), 1N NaOH (18–24 h at 60  C), and 1N HCl again (2–4 h at 60  C) before combustion at 900  C in the presence of cupric oxide and silver foil. The resulting CO2 was purified and converted to graphite in the same manner as above. Conventional radiocarbon ages are reported in 14C yrs and, after calibration, in calendar yrs. For calibration, we used the IntCal09.14C dataset (CALIB 6.0.0, Stuiver and Reimer, 1993; Reimer et al., 2009). 4. Results 4.1. Modern shells A few aliquots (11 of 247, or 4.5% of the total) yielded D14C values that were higher than the modeled dietary D14C values of the corresponding year of collection, which indicates these individuals consumed unusually high amounts of bomb-spike carbon (Table S1, Fig. 6). Data from these shells were excluded from further analysis. For the remaining 236 aliquots of gastropod shells from the 46 different species that we analyzed, w78% did not contain any dead carbon from limestone or other carbonate rocks even though it was readily available at all sites, w12% contained between 5 and 10% dead carbon, and a few (3% of the total) contained more than 10% (Table S1, Fig. 6). D14C values for all taxa ranged from 97.5 to 158.4&, and limestone effects averaged only w180 14C yrs. Dead carbon was not detected in the shells of at least 23 different species, including (number of shells in parentheses) Catinella avara (99), Catinella gelida (66), Catinella vermeta (39), Cochlicopa lubricella (17), Cochlicopa morseana (16), Columella columella (94), Discus catskillensis (40), Discus cronkhitei (47), Discus macclintockii (5), Euconulus alderi (71), Euconulus polygyratus (57), Gastrocopta pentodon (131), Nesovitrea binneyana (46), Punctum minutissimum (542), Pupilla hebes (7), Strobilops affinis (36), Succinea bakeri (27), Succinea grosvernori (1), Succinea n. sp. ‘Minnesota A’ (1), Succinea ovalis (50), Succinea strigata (14), Vertigo hubrichti (144), and Vertigo modesta (73). D14C values were most negative (i.e., contained the most dead carbon) for shells from the Pupilla and Vallonia genera. Maximum 525 limestone effects for these genera ranged from 780  310 14C yrs for P. muscorum to 1590  280 14C yrs for P. sonorana, and from 1010  380 14C yrs for Vallonia perspectiva to 1500  270 14C yrs for V. cyclophorella, respectively (Table 1). The only other species that exhibited a limestone effect that was greater than 1000 14C yrs was H. occulta (1210  250 14C yrs). These results can be applied directly to the fossil record if it is possible to identify the taxa to the species-level based on shell morphology. For the Succineidae family, the data must be evaluated at the family or genus level to be applicable. Taking the Succineidae family as a whole, 85% of the shell aliquots that we analyzed did not contain measurable amounts of dead carbon; the remaining aliquots contained an average of 5.2% dead carbon. Within the family, D14C values for the genus Catinella ranged from 63.7 to 147.0&, Oxyloma values ranged from 0.8 to 135.0&, and Succinea values ranged from 16.1 to 147.8&. Members of the Catinella genus incorporate little, if any, dead carbon from limestone or other carbonate rocks in their shells; 32 of 33 aliquots (97%) of Catinella shells did not contain measurable amounts of dead carbon. The remaining sample (DL-170) contained only a very small amount of dead carbon, equivalent to a limestone effect of 320  310 14C yrs. Similarly, 36 of 39 aliquots (92%) of Succinea shells did not contain dead carbon. The remaining three aliquots were all Succinea indiana; limestone effects for these shells ranged from 430  370 to 610  380 14C yrs. For Oxyloma, 15 of 24 aliquots (63%) did not contain measurable amounts of dead carbon. Limestone effects for the remaining samples ranged between 250  210 and 670  240 14C yrs. 4.2. Modern gastropod bodies We also measured the 14C activity of the bodies of eight gastropods to determine the magnitude of the offset between the body carbon and shell carbonate (Table 2). Ideally, we would have preferred to measure the 14C content of the extrapallial fluid to compare with the shell carbonate to determine if carbon isotopes are fractionated when the shells are formed, but this was not feasible because the gastropod bodies were simply too small. Regardless, the measured D14C values of gastropod body carbon ranged from 42 to 101&, similar to the D14C values of shell carbonate from the same sites, which ranged from 43 to 133&. However, we did not observe a clear relation between the 14C activity of gastropod body carbon and shell carbonate. D14C values of bodies of Oxyloma retusa collected from Maquokata River Mounds, Iowa were indistinguishable from the D14C values of their corresponding shells (D14Cbody ¼ 42  6&, D14Cshell ¼ 43  10&), as were values for S. ovalis from Dave Pepin Homestead, Minnesota (D14Cbody ¼ 101 8&, D14Cshell ¼ 110  8&). In contrast, D14C values for bodies of S. ovalis were significantly lower than their shells from Brewer Boat Ramp, Maine (D14Cbody ¼ 55  6&, D14Cshell ¼ 82  5&) and Zippel Bay State Park, Minnesota (D14Cbody ¼ 84  6&, D14Cshell ¼ 133  12&). The reason(s) for this difference is unclear. 4.3. Fossil shells Well-preserved fossil organic material (bark, plant macrofossils, and wood) recovered from sediments at the Oxford East outcrops yielded calibrated ages that ranged from 24.60  0.40 to 25.28  0.55 ka, with an average of 24.93  0.30 ka (n ¼ 5; Table 3, Fig. 7). Gastropod shells recovered from the same stratigraphic unit yielded ages that ranged from 23.92  0.66 to 25.81  0.94 ka, and averaged 24.73  0.44 ka (n ¼ 18). Average ages of six of the seven fossil taxa were indistinguishable from the average age of the organic matter: Discus shimeki (24.80  0.17 ka; n ¼ 3), P. muscorum (24.34  0.32 ka; n ¼ 3), Vallonia gracilicosta (24.20  0.40 ka; n ¼ 2), Vertigo hannai (24.90  0.77 ka; n ¼ 1), V. modesta 526 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 a Fig. 6. Shell carbonate D14C values compared to modeled dietary D14C values for modern gastropods. Data points that fall on the solid black line in each panel represent gastropods that obtained their carbon from live plants and the atmosphere. Data points that fall below the solid line indicate that dead carbon from limestone or other carbonate rocks was incorporated during shell construction. The magnitude of this phenomenon, called the ‘‘limestone effect’’, depends upon the amount of shell carbon that was derived from rocks as shown by the dashed lines. (24.86  0.32 ka; n ¼ 3), and Succineidae (25.29  0.54 ka; n ¼ 3). The average age of H. occulta (24.34  0.15 ka; n ¼ 3) was slightly younger than the organic ages. 5. Discussion 5.1. Small terrestrial gastropods and the limestone problem Approximately 78% of the modern shells that we analyzed did not contain any dead carbon from limestone or other carbonate rocks even though it was readily available at all sites, w12% of the aliquots contained between 5 and 10% dead carbon, and a few (3% of the total) contained more than 10%. Even at the high end, the amount of dead carbon in the small shells is significantly less than the 20–30% dead carbon that has been previously reported for larger taxa (e.g., Goodfriend and Stipp, 1983). If we extrapolate our results to the fossil record and assume that the shells behave as closed systems with respect to carbon over geologic timescales, then small terrestrial gastropod shells should provide accurate 14C ages w78% of the time and ages that are J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 527 b Fig. 6. (continued). within w1000 14C yrs of the true age w97% of the time. Shells from at least 23 different species did not contain dead carbon, and therefore should yield reliable 14C ages if the modern shell data can be applied directly to the fossil record. For the Succineidae family as a whole, 85% of the shell aliquots that we analyzed did not contain measurable amounts of dead carbon; the remaining aliquots contained an average of 5.2% dead carbon, equivalent to a limestone effect of 425 14C yrs. At the genus level, shells of the genus Catinella should yield reliable 14C ages w97% of the time, again assuming closed-system behavior, and ages that are within w300 14C yrs of the true age every time. Similarly, Succinea shells should yield reliable 14C ages w92% of the time and ages that are within w600 14C yrs of the true age every time. Results for the genus Oxyloma suggest that some caution should be used when evaluating 14C ages derived from these shells. AMS results for the Oxyloma shells show that nearly 1 in 3 aliquots contained at least some dead carbon. Although the amount was relatively minor, <7% of the total, dead carbon was present in Oxyloma shells more frequently than in either Catinella or Succinea shells. 528 Table 1 Summary of J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 14 C results for modern gastropod shells. Limestone effecta (14C yrs) Family Genus Species Aliquots Shells Negligibleb Maximumc Cochlicopidae Cochlicopa Discidae Discus Helicarionidae Euconulus Cochlicopa lubrica Cochlicopa lubricella Cochlicopa morseana Discus catskillensis Discus cronkhitei Discus macclintockii Discus shimeki Euconulus alderi Euconulus dentatus Euconulus fulvus Euconulus polygyratus Hendersonia occulta Punctum minutissimum Columella columella Gastrocopta pentodon Gastrocopta tappaniana Pupilla blandi Pupilla hebes Pupilla muscorum Pupilla sonorana Vertigo elatior Vertigo hannai Vertigo hubrichti Vertigo modesta Vertigo paradoxa Strobilops affinis Strobilops labyrinthica Catinella avara Catinella exile Catinella gelida Catinella vermeta Oxyloma retusa Oxyloma verrilli Succinea bakeri Succinea grosvernori Succinea indiana Succinea n.sp. ‘Minnesota A’ Succinea ovalis Succinea strigata Vallonia cyclophorella Vallonia gracilicosta Vallonia perspectiva Hawaiia minuscula Nesovitrea binneyana Nesovitrea electrina Zonitoides arboreus 3 3 4 6 6 3 7 3 3 8 3 13 3 3 3 3 9 1 8 2 3 3 3 3 3 3 3 9 15 9 4 17 7 8 1 4 1 19 6 6 7 5 3 3 4 4 17 17 16 40 47 5 33 71 34 107 57 13 542 94 131 105 90 7 80 29 113 123 144 73 124 36 43 99 316 66 39 136 47 27 1 6 1 50 14 129 155 221 152 46 33 20 53% 100% 100% 100% 100% 100% 97% 100% 68% 57% 100% 46% 100% 100% 100% 92% 67% 100% 38% 66% 38% 65% 100% 100% 74% 100% 65% 100% 97% 100% 100% 65% 57% 100% 100% 17% 100% 100% 100% 95% 60% 68% 29% 100% 73% 85% 650 – – – – – 430 – 670 730 – 1210 – – – 540 1000 – 780 1590 500 280 – – 380 – 520 – 320 – – 540 670 – – 610 – – – 1500 1370 1010 340 – 710 530 Helicinidae Punctidae Pupillidae Hendersonia Punctum Columella Gastrocopta Pupilla Vertigo Strobilopsidae Strobilops Succineidae Catinella Oxyloma Succinea Valloniidae Vallonia Zonitidae Hawaiia Nesovitrea Zonitoides  390  240  290  300  250  360  270     310 280 280 270  240  320  310  360  240  380     270 270 380 270  240  240 a Defined as the theoretical difference between the measured and true 14C ages for gastropods that incorporate the same amount of dead carbon in their shells as the aliquots measured here. These values are based on the difference between the modeled diet and shell carbonate D14C values and converted into 14C yrs. b Percent of shells measured by AMS that did not contain dead carbon from limestone or other carbonate rocks (i.e., the D14C values for the shells were statistically indistinguishable from the modeled diet D14C value). c Maximum limestone effect for a given taxon (given in 14C yrs). Uncertainties are given at the 2s (95%) confidence level. 5.2. Ca-limiting hypothesis Large gastropod shells (>20 mm in maximum dimension) routinely contain 20–30% dead carbon when living in habitats on carbonate terrain (e.g., Evin et al., 1980; Goodfriend and Stipp, 1983), whereas the small shells measured in this study rarely contained more than w10%. We speculate that calcium may hold the clues to determining the reasons for the difference. Gastropod shell carbonate (aragonite) is composed of three elements – calcium, carbon, and oxygen. The latter two elements are readily available in the environments in which gastropods live and, therefore, cannot be considered as possible limiting factors for shell construction. In most settings, however, calcium is present in plants and water in low concentrations (typically parts per million). If small terrestrial gastropods can acquire enough calcium from their ‘‘normal’’ diet (plants, detritus, and water), then they may not have to consume carbonate rocks to supplement their calcium intake when building their shells. Larger taxa may find it more difficult to obtain enough calcium from these sources without turning to carbonate rocks when they are available. Our results support this hypothesis, but only on a gross scale. There is clearly a significant difference in the amount of dead carbon incorporated in the shells of large taxa previously studied and the small taxa studied here. However, shell size alone is not the only factor to consider when evaluating the results within the small body size class. For example, in the present study, we did not observe a significant correlation between shell size and measured D14C values (R2 ¼ 0.015). The largest taxon that we included in our analysis, H. occulta, averaged 15.6 mg per shell and contained approximately the same amount of dead carbon as P. muscorum and Vallonia shells, which averaged only 1.4 and 0.7 mg per shell, respectively. Similarly, we did not find a clear correlation between shell size and measured D14C values even within a single family. For Succineidae, Catinella shells were generally the smallest, averaging 1.1 mg per shell and contained the least amount of dead carbon, and 529 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 Table 2 14 C results for modern gastropod bodies and corresponding shells. Lab # AA # Taxon Sitea Lat ( N) Long ( W) Mass (mg) d13C (vpdb) F14Cb Shell D14C Atmos D14C Diet D14C Limestone Effectc (14C yrs) Bodies MU-109 MU-114 MU-108 MU-111 MU-117 MU-115 MU-112 MU-113 80177 80181 80176 80178 80183 80182 80179 80180 Succinea ovalis Succinea ovalis Succinea ovalis Succinea ovalis Oxyloma retusa Succinea ovalis Succinea strigata Succinea ovalis 1 2 3 4 5 6 7 8 44.819 48.410 48.906 47.874 42.559 50.264 64.858 48.866 68.723 94.819 96.027 96.422 90.713 66.411 147.862 94.843 4.22 8.64 3.79 11.05 9.04 12.88 4.01 6.88 24.9 24.8 25.1 26.0 26.1 25.0 24.2 23.5 1.0547 1.0996 1.0946 1.0877 1.0419 1.0485 1.0740 1.0833         0.0064 0.0082 0.0062 0.0062 0.0064 0.0072 0.0060 0.0060 55  6 101  8 96  6 89  6 42  6 49  7 74  6 84  6 70 80 80 80 52 57 54 80         5 5 5 5 5 5 5 5 96  33 107  36 107  36 107  36 73  25 79  27 75  26 107  36 330  280 50  300 90  300 150  300 250  210 250  230 0  210 180  300 Shells MU-162 MU-128 MU-179 MU-141 80928 80194 80942 80910 Succinea ovalis Succinea ovalis Oxyloma retusa Succinea ovalis 1 2 5 8 44.819 48.410 42.559 48.866 68.723 94.819 90.713 94.843 8.80 11.72 13.08 8.91 10.0 9.8 11.0 9.6 1.0817 1.1090 1.0425 1.1316     0.0048 82  5 0.0084 110  8 0.0100 43  10 0.0116 133  12 70 80 52 80     5 5 5 5 96  33 107  36 73  25 107  36 110  270 0  300 250  220 0  300 Uncertainties are given at the 2s (95%) confidence level. a Key to sites: 1 ¼ Brewer Boat Ramp, Maine; 2 ¼ Dave Pepin Homestead, Minnesota; 3 ¼ Duxby, Minnesota; 4 ¼ Huot Forest WMA, Minnesota; 5 ¼ Maquokata River Mounds, Iowa; 6 ¼ September Islands, Quebec; 7 ¼ University of Alaska – Fairbanks; 8 ¼ Zippel Bay State Park, Minnesota. b F14 C values are derived from the measured 14C activity, corrected for fractionation, and account for decay that occurred between the time of collection and the AMS measurement. c Defined in Table 1. Oxyloma shells averaged 2.1 mg per shell and contained the most dead carbon. Succinea shells were the largest, averaging 5.8 mg per shell, but were between Catinella and Oxyloma in terms of the amount of dead carbon in their shells (Table S1). The results presented here suggest that the Limestone Problem for small terrestrial gastropods is often negligible and always much less than the 20–30% dead carbon for larger taxa. However, there are additional factors that apparently influence the dietary intake of carbonate rocks of small terrestrial gastropods living side by side, which may include opportunistic behavior, variations in microhabitats, and the dietary needs or wants of individual gastropods. Table 3 14 C results for the Oxford East outcrops. Lab # AA # Taxon Na Organics MU-212 MU-213 MU-214 MU-211 MU-210 82584 82585 82586 82583 82582 bark bark twig wood wood chip Average – – – – – Gastropod shells MU-194 82567 MU-195 82568 MU-196 82569 MU-188 MU-189 MU-190 82561 82562 82563 MU-199 MU-200 MU-201 82572 82573 82574 MU-191 MU-192 MU-193 82564 82565 82566 MU-202 MU-203 82575 82576 MU-205 MU-206 MU-207 MU-208 82577 82578 82579 82580 Discus shimeki Discus shimeki Discus shimeki Average Hendersonia occulta Hendersonia occulta Hendersonia occulta Average Pupilla muscorum Pupilla muscorum Pupilla muscorum Average Succineidae Succineidae Succineidae Average Vallonia gracilicosta Vallonia gracilicosta Average Vertigo hannai Vertigo modesta Vertigo modesta Vertigo modesta Average d13C (vpdb) F14C 4.47 3.28 3.80 3.82 4.29 24.6 25.1 24.9 22.6 23.6 0.0731 0.0767 0.0721 0.0743 0.0772 2 2 2 9.20 10.94 14.99 6.4 6.9 6.4 0.0758  0.0052 0.0762  0.0052 0.0745  0.0053 20.72  0.55 20.68  0.54 20.87  0.57 1 1 1 12.41 10.29 18.42 6.4 6.1 7.1 0.0805  0.0051 0.0787  0.0055 0.0777  0.0055 20.24  0.51 20.42  0.56 20.52  0.57 8 9 10 10.86 11.88 14.13 6.5 6.4 6.4 0.0760  0.0052 0.0805  0.0052 0.0809  0.0051 20.71  0.55 20.24  0.52 20.20  0.51 8 4 15 10.58 9.32 9.53 5.3 5.5 5.8 0.0712  0.0052 0.0689  0.0056 0.0759  0.0055 21.23  0.59 21.49  0.65 20.71  0.58 15 15 10.72 11.08 6.0 6.2 0.0776  0.0058 0.0824  0.0058 20.53  0.60 20.05  0.56 30 15 15 20 10.46 11.47 14.14 15.28 6.5 6.7 7.0 6.7 0.0750 0.0751 0.0771 0.0729 20.81 20.80 20.59 21.04 Mass (mg) 14 C age (ka)          0.0026 0.0025 0.0028 0.0042 0.0025 0.0052 0.0052 0.0053 0.0053 21.02 20.62 21.13 20.88 20.58          0.28 0.27 0.32 0.45 0.26 0.55 0.55 0.55 0.58 Calendar age (ka)b Pc 25.15 24.64 25.28 24.97 24.60 24.93       0.34 0.37 0.55 0.63 0.40 0.30 1.00 1.00 1.00 1.00 1.00 24.73  24.67  25.00  24.80  24.18  24.37  24.47  24.34  24.71  24.18  24.13  24.34  25.33  25.81  24.73  25.29  24.49  23.92  24.20  24.90  24.88  24.54  25.18  24.86  0.75 0.74 0.79 0.17 0.67 0.68 0.70 0.15 0.74 0.68 0.67 0.32 0.80 0.94 0.78 0.54 0.76 0.66 0.40 0.77 0.77 0.71 0.77 0.32 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.98 1.00 1.00 1.00 1.00 Uncertainties for the raw and calibrated 14C ages are given at the 2s (95%) confidence level. a Number of shells per aliquot. b Calibrated ages were calculated using CALIB v. 6.0.0, IntCal09.14C dataset; limit 50.0 calendar ka B P. Calibrated ages are reported as the midpoint of the calibrated range. Uncertainties are reported as the difference between the midpoint and either the upper or lower limit of the calibrated age range, whichever is greater. Multiple ages are reported when the probability of a calibrated age range exceeds 0.05. c P ¼ probability of the calibrated age falling within the reported range as calculated by CALIB. 530 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 Fig. 7. Photograph of the section at the Oxford East outcrops and the calibrated ages obtained from the organic material and fossil gastropod shells. 5.3. Small terrestrial gastropods and open-system behavior The present study includes only a single fossil locality, the Oxford East outcrops of southwestern Ohio, which we present here as a case study. Average ages of organic materials and fossil gastropod shells from the Oxford East outcrops were statistically indistinguishable; 24.93  0.30 ka for the organics and 24.73  0.44 ka for the shells. Six of the seven taxa that we analyzed (D. shimeki, P. muscorum, Succineidae, V. gracilicosta, V. hannai, and V. modesta) yielded average 14C ages that are indistinguishable from the organic ages; the seventh (H. occulta) yielded ages that were slightly younger than the organic ages. The range of ages of the shell material, 1.8 ka, is significantly larger than the range of ages of the organics, 0.8 ka. If the dispersion of shell ages was related to opensystem behavior, then we would expect the ages to be systematically younger than the organic matter ages, which they are not. It may be that the small number of organic samples fails to adequately capture the full range of time represented by the sampled stratum. More work needs to be done on this subject, including additional comparisons of shell and organic ages from other sites, but the results from the Oxford East outcrops suggest that small terrestrial gastropod shells may behave as closed systems with respect to carbon in the American Midwest for at least the past w25 ka. 6. Summary and conclusions Fossil shells of small terrestrial gastropods are commonly preserved in Quaternary sediment across North America, including loess, wetland, glacial, and alluvial deposits, as well as in sediments at many archeological sites. Their aragonitic shells contain w12% by weight carbon, and therefore contain sufficient carbon for 14C dating. However, terrestrial gastropod shells in carbonate terrains are often avoided for 14C dating because large taxa are known to incorporate dead carbon from limestone or other carbonate rocks when building their shells, which can cause their 14C ages to be up to 3000 yrs too old. Previous studies suggested that small terrestrial gastropod shells may yield reliable 14C ages in arid environments, but a systematic and comprehensive analysis was needed before ages derived from their shells could be considered reliable outside of the Desert Southwest. To this end, we measured the 14C activity of 247 aliquots of modern shell material (3749 individual shells) from 163 localities across North America. Approximately 78% of the aliquots did not contain measurable amounts of dead carbon even though limestone or other carbonate rocks were readily available at all sites, w12 of the aliquots contained between 5 and 10% dead carbon, and the remaining few (3% of the total) contained more than 10%. The average Limestone Effect for these samples was only w180 14C yrs, which is significantly less than the 2000–3000 14C yrs that previous researchers found for larger taxa. Assuming that the small gastropod shells behave as closed systems with respect to carbon after burial, they should yield reliable 14C ages w78% of the time, and ages that are within w1000 yrs of the true age w97% of the time, regardless of the taxon analyzed, local bedrock type, climate, or environmental conditions. If fossil shells can be identified to the species level, then at least 23 different species should yield reliable 14C ages if the modern shell data can be applied directly to the fossil record. The terrestrial gastropod family Succineidae is one of the most common gastropod taxa in North America. Unlike the other J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 gastropods studied here, our 14C data for Succineidae must be evaluated at the genus or even family level because species-level identification is based on soft-part morphology, which is rarely preserved in the fossil record. Based on the data from modern shells, the Succineidae family as a whole should yield reliable 14C ages w85% of the time, and ages that are within w700 14C yrs every time. At the genus level, Catinella should yield reliable 14C ages w97% of the time, again assuming closed-system behavior, and ages that are within w300 14C yrs of the true age every time. Similarly, Succinea shells should yield reliable 14C ages w92% of the time and ages that are within w600 14C yrs of the true age every time. Caution should be used when evaluating shells of the genus Oxyloma, however, as nearly 1 in 3 aliquots contained dead carbon, equivalent to a limestone effect of up to w700 14C yrs. Fossil shells of small terrestrial gastropods recovered from welldated, late-Pleistocene sediments in the Midwest yielded ages that were statistically indistinguishable from ages obtained from wellpreserved plant macrofossils (wood, bark, plant remains). Although just one site, these results suggest that small terrestrial gastropod shells may behave as closed systems with respect to carbon over geologic timescales. More work on this subject is needed, but if our case study site is representative of other sites, then fossil shells of some small terrestrial gastropods, including at least five common genera, Catinella, Columella, Discus, Gastrocopta, and Succinea, should yield reliable 14C ages, regardless of the local geologic substrate. Fossil shells of these and other small terrestrial gastropods are common in a wide range of Quaternary deposits in North America and, therefore, our results may have broad chronologic applications to Quaternary geology and New World archeology. Acknowledgments We thank K. Gotter for converting carbon dioxide samples to graphite at the JAR’s lab at Miami University and J. Quade and the University of Arizona Desert Laboratory for access to their vacuum extraction systems. We also thank J. Rosenbaum and D. Van Sistine of the U.S. Geological Survey for their help with Figs. 1 and 2, respectively. This manuscript benefited from constructive reviews from E. Ellis, G. Hodgins, D. Muhs, M. Reheis, J. Southon, and two anonymous reviewers. This project was funded by the National Science Foundation Sedimentary Geology and Paleobiology Competition, award #EAR 0614840. Additional support was provided to JSP by the U.S. Geological Survey’s Mendenhall postdoctoral research program. Appendix. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.quageo.2010.01.001. Editorial Handling by: A. Hogg References Balakrishnan, M., Yapp, C.J., 2004. Flux balance models for the oxygen and carbon isotope compositions of land snail shells. Geochimica et Cosmochimica Acta 68, 2007–2024. Barker, G.M., 2001. Gastropods on land: phylogeny, diversity, and adaptive morphology. In: Barker, G.M. (Ed.), Biology of Terrestrial Molluscs. CABI Publishing, Oxon, U.K, pp. 1–146. Brennan, R., Quade, J., 1997. Reliable Late-Pleistocene stratigraphic ages and shorter groundwater travel times from 14C in fossil snails from the southern Great Basin. Quaternary Research 47, 329–336. Brovkin, V., Cherkinsky, A., Goryachkin, S., 2008. Estimating soil carbon turnover using radiocarbon data: a case study for European Russia. Ecological Modelling 216, 178–187. 531 Burch, J.B., Pearce, T.A., 1990. Terrestrial Gastropoda. Soil Biology Guide. John Wiley & Sons, New York. Cameron, R.A.D., Pokryszko, B.M., 2005. Estimating the species richness and composition of land mollusc communities. Journal of Conchology 38, 529–547. Dell, A.M., 1991. Reconstruction of late Pleistocene paleoecology in southwestern Ohio from nonmarine gastropod assemblages. Unpublished M.S., University of Cincinnati, 97 p. Eckberg, M.P., Lowell, T.V., Stuckenrath, R., 1993. Late Wisconsin glacial advance and retreat patterns in southwestern Ohio, USA. Boreas 22, 189–204. Emberton, K.C., Pearce, T.A., Randalana, R., 1996. Quantitatively sampling land-snail species richness in Madagascan rainforests. Malacologia 38, 203–212. Evans, J.G., 1972. Land Snails in Archaeology. Seminar Press, New York, NY. Evin, J., Marechal, J., Pachiaudi, C., 1980. Conditions involved in dating terrestrial shells. Radiocarbon 22, 545–555. Frye, J.C., Willman, H.B., 1960. Classification of the Wisconsinan stage in the Lake Michigan glacial lobe. Illinois Geological Survey Circular 285, 1–16. Goodfriend, G.A., 1987. Radiocarbon age anomalies in shell carbonate of land snails from semi-arid areas. Radiocarbon 29, 159–167. Goodfriend, G.A., Hood, D.G., 1983. Carbon isotope analysis of land snail shells: implications for carbon sources and radiocarbon dating. Radiocarbon 25, 810–830. Goodfriend, G.A., Stipp, J.J., 1983. Limestone and the problem of radiocarbon dating of land-snail shell carbonate. Geology 11, 575–577. Goslar, T., Pazdur, M.F., 1985. Contamination studies on mollusk shell samples. Radiocarbon 27, 33–42. Grosjean, M., 1994. Paleohydrology of the Laguna Lejı́a (North Chilean Altiplano) and climatic implications for late-glacial times. Palaeogeography, Palaeoclimatology, Palaeoecology 109, 89–100. Guadinski, J.B., Trumbore, S.E., Davidson, E.E., Zheng, S., 2000. Soil carbon cycling in a temperate forest: radiocarbon-based estimates of residence times, sequestration rates, and partitioning of fluxes. Biogeochemistry 51, 33–69. Hsueh, D.Y., Karkauer, N.Y., Randerson, J.T., Xu, X., Trumbore, S.E., Southon, J.R., 2007. Regional patterns of radiocarbon and fossil fuel-derived CO2 in surface air across North America. Geophysical Research Letters 34, L02816. doi:10.1029/ 2006GL027032. Hua, Q., 2004. Review of tropospheric bomb 14C data for carbon cycle modeling and age calibration purposes. Radiocarbon 46, 1273–1298. Hubricht, L., 1985. The distributions of the native land mollusks of the eastern United States. Fieldiana: Zoology 24, 1–191. Kerney, M., Cameron, R.A.D., 1979. Field Guide to the Land Snails of Britain and Northwestern Europe. Collins Press, London. Leighton, M.M., 1960. The classification of the Wisconsin Glacial stage of north central United States. Journal of Geology 68, 529–552. Lowell, T.V., 1995. The Application of Radiocarbon Age Estimates to the Dating of Glacial Sequences: an Example from the Miami Sublobe, Ohio, USA. Quaternary Science Reviews 4, 85–99. Manning, M.R., Lowe, D.C., Melhuish, W.H., Sparks, R.J., Wallace, G., Brenninkmeijer, C.A.M., McGill, R.C., 1990. The use of radiocarbon measurements in atmospheric studies. Radiocarbon 32, 37–58. Meijer, H.A.J., van der Plicht, J., Gislefoss, J.S., Nydal, R., 1995. Comparing long term atmospheric 14C and 3H records near Groningen, the Netherlands with Fruholmen, Norway and Izaña, Canary Islands 14C stations. Radiocarbon 37, 39–50. Nekola, J.C., 2004. Terrestrial gastropod fauna of northeastern Wisconsin and the southern Upper Peninsula of Michigan. American Malacological Bulletin 18, 21–44. Nekola, J.C., 2005. Geographic variation in richness and shell size of eastern North American land snail communities. Records of the Western Australian Museum Supplement 68, 39–51. Oggier, P., Zschokke, S., Baur, B., 1998. A comparison of three methods for assessing the gastropod community in dry grasslands. Pedobiologia 42, 348–357. Pedone, V., Rivera, K., 2003. Groundwater-discharge deposits in Fenner Wash, eastern Mojave Desert. Geological Society of America Abstracts with Programs 35, 257. Pigati, J.S., Bright, J.E., Shanahan, T.M., Mahan, S.A., 2009. Late Pleistocene Paleohydrology Near the Boundary of the Sonoran and Chihuahuan Deserts, Southeastern Arizona, USA. Quaternary Science Reviews 28, 286–300. Pigati, J.S., Quade, J., Shanahan, T.M., Haynes Jr., C.V., 2004. Radiocarbon dating of minute gastropods and new constraints on the timing of spring-discharge deposits in southern Arizona, USA. Palaeogeography, Palaeoclimatology, Palaeoecology 204, 33–45. Placzek, C., Quade, J., Patchett, P.J., 2006. Geochronology and stratigraphy of late Pleistocene lake cycles on the southern Bolivian Altiplano: implications for causes of tropical climate change. Geological Society of America Bulletin 118, 515–532. Ponder, W.F., Lindberg, D.R., 1997. Towards a phylogeny of gastropod molluscs: an analysis using morphological characters. Zoological Journal of the Linnean Society 119, 83–265. Quarta, G., Romaniello, L., D’Elia, M., Mastronuzzi, G., Calcagnile, L., 2007. Radiocarbon age anomalies in pre- and post-bomb land snails from the coastal Mediterranean basin. Radiocarbon 49, 817–826. Reimer, P., Baillie, M., Bard, E., Bayliss, A., Beck, J., Blackwell, P., Ramsey, C.B., Buck, C., Burr, G., Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Hajdas, I., Heaton, T., Hogg, A., Hughen, K., Kaiser, K., Kromer, B., McCormac, F., Manning, S., Reimer, R., Richards, D., Southon, J., Talamo, S., Turney, C., van der Plicht, J., Weyhenmeyer, C., 2009. IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal B.P. Radiocarbon 51, 1111–1150. 532 J.S. Pigati et al. / Quaternary Geochronology 5 (2010) 519–532 Reimer, P.J., Brown, T.A., Reimer, R.W., 2004. Discussion: reporting and calibration of post-bomb 14C data. Radiocarbon 46, 1299–1304. Riggs, A.C., 1984. Major carbon-14 deficiency in modern snail shells from southern Nevada springs. Science 224, 58–61. Romaniello, L., Quarta, G., Mastronuzzi, G., D’Elia, M., Calcagnile, L., 2008. 14C age anomalies in modern land snails shell carbonate from Southern Italy. Quaternary Geochronology 3, 68–75. Rubin, M., Likins, R.C., Berry, E.G., 1963. On the validity of radiocarbon dates from snail shells. Journal of Geology 71, 84–89. Samos, G., 1949. Some observations on exchange of CO2 between BaCO3 and CO2 gas. Science 110, 663–665. Slota, P.J., Jull, A.J.T., Linick, T.W., Toolin, L.J., 1987. Preparation of small samples for 14 C accelerator targets by catalytic reduction of CO. Radiocarbon 29, 303–306. Stott, L.D., 2002. The influence of diet on the d13C of shell carbon in the pulmonate snail Helix aspersa. Earth and Planetary Science Letters 195, 249–259. Stuiver, M., Polach, H.A., 1977. Reporting of 14C data. Radiocarbon 19, 355–363. Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon 35, 215–230. Tamers, M.A., 1970. Validity of radiocarbon dates on terrestrial snail shells. American Antiquity 35, 94–100. Torn, M.S., Vitousek, P.M., Trumbore, S.E., 2005. The influence of nutrient availability on soil organic matter turnover estimated by incubations and radiocarbon modeling. Ecosystems 8, 352–372. Wilbur, K.M., 1972. Shell formation in mollusks. In: Florkin, M., Scheer, B.T. (Eds.), Chemical Zoology, Mollusca, vol. 7. Academic Press, New York, pp. 103–145. Yates, T., 1986. Studies of non-marine mollusks for the selection of shell samples for radiocarbon dating. Radiocarbon 28, 457–463. Zhou, W., Head, W.J., Wang, F., Donahue, D.J., Jull, A.J.T., 1999. The reliability of AMS radiocarbon dating of shells from China. Radiocarbon 41, 17–24.