Palaeontologia Electronica palaeo-electronica.org

Methods for isolation and quantification of microfossil fish teeth and elasmobranch dermal denticles (ichthyoliths)

from marine sediments

Elizabeth C. Sibert, Katie L. Cramer, Philip A. Hastings, and Richard D. Norris

ABSTRACT

Ichthyoliths—microfossil fish teeth and shark dermal scales (denticles)—arefound in nearly all marine sediments. Their small size and relative rarity compared toother microfossil groups means that they have been largely ignored by the paleontol-ogy and paleoceanography communities, except as carriers of certain isotopic sys-tems. Yet, when properly concentrated, ichthyoliths are sufficiently abundant to revealpatterns of fish abundance and diversity at unprecedented temporal and spatial resolu-tion, in contrast to the typical millions of years-long gaps in the vertebrate body fossilrecord. In addition, ichthyoliths are highly resistant to dissolution, making it possible toreconstruct whole fish communities over highly precise and virtually continuous times-cales. Here we present methods to isolate and utilize ichthyoliths preserved in the sed-imentary record to track fish community structure and ecosystem productivity throughgeological and historical time periods. These include techniques for isolation and con-centration of these microfossils from a wide range of sediments, including deep-seaand coral reef carbonates, clays, shales, and silicate-rich sediments. We also presenta novel protocol for ichthyolith staining using Alizarin Red S to easily visualize and dis-tinguish small teeth from debris in the sample. Finally, we discuss several metrics forquantification of ichthyolith community structure and abundance, and their applicationsto reconstruction of ancient marine food webs and environments.

Elizabeth C. Sibert. Society of Fellows, Harvard University, 78 Mount Auburn Street, Cambridge, Massachusetts 02138, USA; Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Drive MC 0208, La Jolla, California 92093, USA. esibert@fas.harvard.edu Katie L. Cramer. Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Drive MC 0208, La Jolla, California 92093, USA. kcramer@ucsd.edu Philip A. Hastings. Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Drive MC 0208, La Jolla, California 92093, USA. phastings@ucsd.edu Richard D. Norris. Scripps Institution of Oceanography, University of California, San Diego, 9500 Gilman Drive MC 0244, La Jolla, California 92093, USA. rnorris@ucsd.edu

Keywords: fish teeth; denticles; ichthyoliths; coral reefs; historical ecology; micropaleontology

Submission: 4 May 2016 Acceptance: 13 March 2017

Sibert, Elizabeth C., Cramer, Katie L., Hastings, Philip A., and Norris, Richard D. 2017. Methods for isolation and quantification of microfossil fish teeth and elasmobranch dermal denticles (ichthyoliths) from marine sediments. Palaeontologia Electronica 20.1.2T: 1-14palaeo-electronica.org/content/2017/1800-quantifying-ichthyoliths

Copyright: April 2017 Palaeontological Association

SIBERT: QUANTIFYING ICHTHYOLITHS

INTRODUCTION

Despite being relatively common in marinesediments, ichthyoliths, the isolated microfossilteeth, dermal denticles, and bone fragments ofsharks and bony fishes (Figure 1), have been over-looked by much of the scientific community, over-shadowed by more abundant and better studiedforaminifera, nannofossils, and other microfossils,for research into biological responses to ancientclimate and environmental change (Cifelli, 1969;Frerichs, 1971; Smit, 1982; Hallock and Schlager,1986; Kelly et al., 1998; Hull et al., 2011). Whileunderstanding the response of these unicellularorganisms to climate and biotic events providesinsight into the sensitivity of marine ecosystems toglobal change, unicellular algae and protists areonly the base of a complex marine ecosystem,which support a diverse array of consumers,including marine vertebrates.

Fishes are one of the most diverse and eco-logically successful vertebrate clades (Nelson,2006; Friedman and Sallan, 2012; Near et al.,2013) and are a hallmark of nearly all marine eco-systems. The presence and abundance of fish bio-mass is an indicator of how efficiently anecosystem is functioning, in terms of transferringenergy from the base of the food web to the uppertiers (Sprules and Munawar, 1986; Iverson, 1990),

and therefore the abundance and composition ofichthyoliths may be a reasonable proxy for ecosys-tem structure and function. Moreover, there aretypically excellent chronologies and relatively con-tinuous sedimentation rates in many deep-sea sed-imentary sequences (e.g., Hilgen, 1991;Westerhold et al., 2008; Hilgen et al., 2010). Thus,it is possible to capture unusually detailed historiesof vertebrates, as compared to the typical temporaland spatial fragmentation of the terrestriallyexposed body-fossil record. In recent, shallowmarine sediments, ichthyoliths can reveal changesin both diversity and abundance of fishes andsharks in coastal systems—making it possible toreconstruct fish community responses to overfish-ing, reef environmental decline and anthropogenicclimate change (Jackson et al., 2001).

While Paleozoic ichthyoliths have a rich his-tory of study (Maisey, 1984; Turner, 2004), youngerichthyoliths (Late Mesozoic and Cenozoic) havelargely been ignored by the paleontology commu-nity, excepting large shark teeth (Cappetta andSchultze, 2012), as the majority of stem diversityfor living clades was established by the Mesozoic.Cretaceous and Cenozoic ichthyoliths have beenused as carriers of several isotopic proxies, includ-ing neodymium as a water-mass tracer (Martin andHaley, 2000; Scher and Martin, 2004), and stron-tium, which can be used both as a weathering

FIGURE 1. An assortment of large (>106 μm fraction) denticles (elasmobranch scales; left) and fish teeth (right) fromDSDP Site 596, a red clay core in the South Pacific. These ichthyoliths are approximately 52 million years old. Imagewas taken on the Hull Lab Imaging System, Yale University. Scale bar is 500 μm.

2

PALAEO-ELECTRONICA.ORG

proxy, and for rough dating of sediments (Ingram,1995; Gleason et al., 2002, 2004, 2008). The fieldof ichthyolith biostratigraphy was developed in theearly 1970s (Doyle et al., 1974; Edgerton et al.,1977). Composed of calcium-phosphate, ichthyo-liths are extremely dissolution resistant, and areone of the last microfossil groups remaining inmarine sediments exposed to corrosive bottomwater and can be used to date fossil-poor pelagicred clays (Doyle and Riedel, 1979b, 1985; Doyle etal., 1988). An updated ichthyolith biostratigraphyfor the Eastern North Pacific was developed in2006 (Johns et al., 2005, 2006).

However, the Cretaceous and Cenozoic ich-thyolith record can reveal important informationabout the role of fishes in aquatic ecosystems andtheir response to global change events (Sibert etal., 2014, 2016; Sibert and Norris, 2015). As ichthy-oliths are found in nearly all sediment types, includ-ing those of the open ocean, which are rarelypreserved on land, pelagic ichthyoliths represent afossil record virtually untouched by traditionalpaleoichthyology. Further, Holocene ichthyolithrecords have the potential for identification toextant taxa, and can show changes in functionaland taxonomic groups over prehistorical and his-torical time periods resulting from environmentaland/or anthropogenic change (Cramer et al.,2017). For example, on modern coral reefs, theabundance of coral-associated fishes is a reliableindicator of coral abundance and growth, andintensive algal grazing by herbivorous fishes facili-tates coral dominance (Randall, 1961; Bellwoodand Wainwright, 2002). Thus the ichthyolith record,in conjunction with other microfossil and geochemi-cal records, can provide insight into ecosystemresponse and resilience to climatic, biotic, andeven anthropogenic perturbations (Sibert et al.,2014; Sibert and Norris, 2015; Cramer et al.,2017). Lastly, understanding how this group of con-sumers has responded to global change eventsmay also yield insights into the mechanismsbehind Cenozoic marine vertebrate evolution andthe development of the vast diversity of fish clades(Nelson, 2006; Near et al., 2012, 2013; Betancur-Ret al., 2013; Broughton et al., 2013).

Here, we provide a detailed methodologicalframework for the isolation, concentration, andanalysis of ichthyoliths as a paleoceanographic,paleoecological, and paleontological resource.Although the methods presented here have beendeveloped and tested with deep-sea sedimentsand near-modern coral reef sediments, we believe

that they can be translated to other marine andlacustrine sediments as well.

METHODS FOR ICHTHYOLITH ISOLATION AND CONCENTRATION

We present methods for isolation of ichthyo-liths from a variety of sediment types, summarizedin a flow chart (Figure 2), and discuss the specificsof each protocol within the text. It is usually imprac-tical to sort through disaggregated sediments forichthyoliths due to their small size and rarity com-pared to other microfossils such as benthic andplanktonic foraminifera and other coarse-grainedsediment clasts. Because metrics of ichthyolithaccumulation (abundance) and community struc-ture rely on the quantification of all ichthyoliths in asample, as opposed to a randomly sampled sub-set, it is necessary to concentrate the full ichthyo-lith assemblage from a raw sediment sample.Processing a sediment sample for ichthyoliths is abalance between efficient concentration (typicallyby disaggregation of sediment and washingthrough a fine sieve), and minimization of potentialloss of teeth by dissolution, fragmentation oradherence onto surfaces such as paintbrushes,splitters, vials, or other surfaces during processingand picking. While ichthyoliths are composed ofcalcium phosphate (bio-apatite), which is resistantto dissolution, care must be taken to counteractpotential destruction and loss of ichthyoliths whenusing methods of acid preparation or bleach-medi-ated disaggregation of sediments. Once washed,ichthyoliths are picked out of the remaining residueusing a high-power dissection microscope andextremely fine paintbrush.

A challenge in working with ichthyoliths is theirsmall size: the vast majority of teeth in pelagic sed-iments are only retained on a 38 μm screen, pass-ing through the typical 63 μm sieves used for mostforaminifera work. Modern reef fish teeth aresomewhat larger, retained on 63 µm screens, how-ever they are some of the smallest-sized compo-nents of reef sediments. As a practical matter, mostpelagic fish teeth are conical or triangular, and willslip through the larger 63μm sieve even if they aremuch longer than 63 μm. We have found thatupwards of 50–80% of the total ichthyolith assem-blage in pelagic sediments is represented by the38–63 μm fraction in pelagic sediments. It is likelythat using a sieve smaller than 38 µm would yieldadditional ichthyoliths, as the majority of teeth inour samples are in the 38–63 μm fraction, howeverthe <38 μm fraction presents significant technical

3

SIBERT: QUANTIFYING ICHTHYOLITHS

challenges for reflected-light microscope-basedwork.

Carbonates

Acid-resistant calcium phosphate ichthyolithscan be extracted from marine carbonates by aceticacid dissolution of the calcium carbonate matrix. Indeep-sea sediments, carbonate-hosted ichthyolithassemblages can be placed on the highly resolvedtime scales derived from analysis of other micro-fossil groups, magnetic reversals, or astrochronol-ogies (e.g., Sibert et al., 2014). In Holocene coralreef sediments, high-precision uranium-thoriumdating of coral skeletons can provide extremelywell-resolved chronologies of fish communitiesover prehistorical and historical time (Cramer et al.,2017). These precise time scales provide esti-mates of sedimentation rate and mass accumula-tion rate, which can be used to estimate fish and

elasmobranch abundance or productivity. Com-bined with the rich abundance of microfossil plank-ton, or the diverse hard parts of reef-associatedanimals, foraminifera and algae, a well-studied car-bonate section can yield information about manycomponents of an ecosystem through an interval ofinterest, giving environmental and ecological con-text to an ichthyolith record (Sibert et al., 2014;Cramer et al., 2017).Deep-sea carbonate ooze and chalk. Simplypicking ichthyoliths out of the coarse fraction of car-bonate sediments is time consuming, and oftenleads to poor data quality, as the small teeth areobscured by the high abundance of foraminiferaand siliceous microfossils. To concentrate ichthyo-liths effectively and address these issues, samplesare dried to a constant weight and then dissolvedin 5–10% acetic acid. Acid is added to the samplesin 100–200 ml increments and stirred every ~20–

FIGURE 2. A flowchart showing the steps for sediment processing for efficient and effective ichthyolith isolation from avariety of sediment types. Sediment types are in boxes, while processing steps are shown in ovals.

4

PALAEO-ELECTRONICA.ORG

30 minutes, until no carbonate remains, usuallyafter ~2–5 hours. The reaction is considered com-plete when no bubbles are released when addingacid or stirring. We find that between 30 and 80 mlof dilute acetic acid is needed per gram of dry sedi-ment to completely dissolve all of the carbonate ina sample, depending both on the concentration ofacid used and the percent-carbonate compositionof the sediments. We do not observe any etchingor other damage to ichthyoliths during this process,and indeed, due to their high abundance andexceptional preservation in red clays, which wereexposed to millions of years of relatively corrosivebottom water, it is unlikely that this limited exposureto weak acetic acid damages the ichthyoliths.Indeed, time-series experiments comparing aceticacid dissolution to hydrochloric acid dissolutionshow that while HCl causes significant degradationto ichthyoliths within 30 minutes of exposure, ace-tic acid does not harm the ichthyoliths even afterprolonged soaking. However, to avoid any potentialdestruction of ichthyoliths, acid exposure should belimited and dilute acetic acid should be used. Oncedissolved, the sample is washed over a 38 μmscreen, and the residue is transferred to filter paperin a funnel and dried in a 50°C oven.

Although it is destructive of the calcareousfossils, dissolution of bulk carbonate samples forichthyoliths as outlined above, is by far the mosteffective method for ichthyolith work. It also yieldsthe highest data quality, as every transfer of thesample between containers leads to some loss ofichthyoliths. Bulk dissolution followed by washingalso uses the least amount of water per sample.However, if it is imperative to preserve certain car-bonate microfossils in a sample, such as largerbenthic foraminifera for isotope studies, or foramin-ifera from a critical interval, we suggest a double-washing procedure: the first wash is carried outwith de-ionized water only, to retain the coarsefraction of carbonates >38 µm. All material below aspecific size threshold (e.g., 150 μm or 250 μm,study-specific) is then dissolved to concentrate thesmaller ichthyoliths. We have found that ichthyo-liths are selectively lost in sample splitters due tostatic adhesion and recommend against their use.As the volume of coarse-grained carbonate sedi-ment is relatively small, it is feasible, although timeconsuming, to pick out all of the ichthyoliths in the>150 µm or >250 µm calcareous residues. In thiscase, it is most important that the processingmethod be internally consistent for an entire sam-ple set, and the potential biases recognized whencomparing absolute ichthyolith abundance values

to other records. Additionally, large teeth (>150μm) are relatively rare in pelagic sediments and, ifignored or under-counted, will not greatly bias thetotal ichthyolith accumulation rate. Indeed, it is alsopossible to count only the fraction subjected to acidtreatment since the fine fractions retain the vastmajority of teeth in a given sample; however in thiscase, information on the maximum size of teeth, orthe change in abundance of specific large teethand denticles, which are almost exclusively >100μm will be missed, which can preclude significantbiological findings (Sibert and Norris, 2015). There-fore, the exact method employed will depend onthe goals of the study. Above all, it is important tomaintain consistency in processing methodthroughout the entire record. Limestones. Lithified limestone also yields ichthy-oliths in the acid-insoluble fraction, however, theprocessing is slightly different from that used fordeep-sea carbonate sediments. Limestones shouldbe broken up into ~1 cm pieces, to increase sur-face area exposed to acid, while preserving themicrofossils – in our experience, shatterboxes andother crushing tools can damage the fossils. Ourapproach comprises barely covering samples with10% acetic acid—we have found that 5% is ineffec-tive for the majority of limestones—and changingthe acid bath every 24 hours. When changing theacid, the sample is washed over a stack of sieveswith all pieces of limestone >150 μm returned tofresh acid, and all residues 38–150 μm areretained to pick through for ichthyoliths. The pro-cess takes approximately 5–12 washes, dependingon the degree of lithification of the rock and thesize of the original limestone fragments (Sibert etal., 2014). Coral reef sediments. Modern reef sediments arecomprised almost entirely of carbonate grains fromcalcifying organisms including corals, mollusks,echinoderms, foraminifera, calcareous algae,sponges, and crustaceans. To preserve theseother taxonomic groups, which are mostly >500μm, only the fraction <500 μm of these cores isdigested in acid and picked through for ichthyoliths,the majority of which are <250 μm. Due to the sandto pebble-sized carbonate grains in reef sediments,10% acetic acid is required. Two to four applica-tions of approximately 200 ml of acid are added todry, size-fractioned sediments every 24 hours.When gentle stirring of acid and residues fails tocause further reaction and residues darken due todominance of organic material following eliminationof carbonates, samples are transferred to a 63 μmsieve and washed with DI water until the water

5

SIBERT: QUANTIFYING ICHTHYOLITHS

leaving the sieve runs clear. Coral reef sedimentsoften have considerable organic matter and debris,which causes excessive clumping of ichthyolithsand highly abundant but acid-insoluble siliceoussponge spicules. To remove excess organic matter,the washed samples can be treated with 25 mldilute chlorine bleach poured directly onto thesieve following an initial wash, left for approxi-mately 1 minute, then rinsed with DI water andlightly agitated until water runs clear below thesieve. The samples are then transferred to filterpaper in a funnel and dried at 50°C. In contrast topelagic sediments, where the majority of ichthyo-liths are within the 38–63 μm fraction, the vastmajority of ichthyoliths preserved in reef sedimentsare >63 μm, so the 63 μm sieve size is used tofacilitate washing the larger initial sample volumesnecessary in these high sedimentation rate sys-tems (Cramer et al., 2017).

Pelagic Clays

Pelagic clays yield, by far, the greatest abun-dance of ichthyoliths per gram sediment: the slowsedimentation rate below the carbonate compen-sation depth, and small grain size means that ich-thyoliths are highly concentrated, and are typicallywell-preserved in these sediments. However, theslow sedimentation rate and lack of other biostrati-graphically well-calibrated microfossils such as cal-careous nannofossils, mean that clays often havepoor age constraints, and there may be very littlepaleoenvironmental context from traditional prox-ies.

To isolate ichthyoliths from pelagic clay, thesamples are dried completely to enable the calcu-lation of ichthyolith accumulation rates. We havefound that many pelagic clay samples fail toachieve stable dry weights for many weeks, per-haps because of water bound in clays. However,once the samples are completely dried, they aresimply disaggregated in de-ionized water, washedover a 38μm sieve, transferred to filter paper in afunnel, dried in a 50°C oven, and examined underthe microscope. These residues, in the best cir-cumstance, may contain only fish teeth and dermaldenticles, however, in some cases may also con-tain micro-Manganese nodules, siliceous microfos-sils, terrigenous sediment clasts, zeolites, orclumps of Fe-oxides.

Silica-dominated Sediments

Siliceous sediments, whether from quartz siltor biogenic opal, create a distinctive challenge inthe isolation and quantification of ichthyoliths. Sil-

ica is insoluble in acetic acid and thus increasesthe volume of the acid-insoluble coarse-fractioncontaining ichthyoliths. Additionally, many quartzgrains have a significant visual similarity to tiny ich-thyoliths at first glance, making visual identificationand picking a challenge. We have found two meth-ods to be effective in isolation of teeth in siliceoussediments—the use of Alizarin Red S, a calcium-specific stain to color ichthyoliths and make themvisible against a backdrop of translucent silica. Forparticularly challenging samples, heavy liquid sep-aration may remove most of the low density sili-ceous sediment relative to ichthyoliths, however,we do not recommend this as a first line of ichthyo-lith concentration methods.Alizarin Red S. Visual differentiation of fish teethfrom other small triangular sediment grains canoften be confounded at small size fractions (<63μm). However, ichthyoliths can be preferentiallystained by Alizarin Red S (1,2-dihydroxyanthraqui-none, C14H8O4), a calcium-specific dye commonlyused in clearing-and-staining fish (Taylor, 1967;Song and Parenti, 1995). Alizarin Red S gives ich-thyoliths a pink tinge, while leaving the silica grainsuntouched (Figure 3). Alizarin is a pH sensitivedye, which turns a deep purple in basic solution,and when in contact with calcium, will adhere to it,leaving a pink or red color. Alizarin Red S is not apanacea: it will also dye calcium carbonate grains,and thus is used most effectively after the carbon-ate fraction has been removed from a sample viaacid dissolution (see Figure 2). However we havefound that even in samples with carbonate, ichthy-oliths are preferentially stained a more intensecolor than foraminifera. We modified a clearing-and-staining protocol for fishes, based on both apublished protocol (step 9, Song and Parenti,1995), and the protocol used by Scripps MarineVertebrate Collection, to use a 1% potassiumhydroxide (KOH) solution with enough Alizarin toturn the solution a deep purple (a surprisingly smallamount). The KOH/Alizarin Red S solution isadded to the post-acid, washed and dried residue.The volume of Alizarin + KOH solution needed isdependent on the amount of residue: generally,just a few drops of the solution, enough to coverthe sample residue in its container, is more thansufficient to produce the desired effect. This is leftfor 24 to 48 hours, and then washed over a 38 μmscreen, transferred to filter paper in a funnel, anddried overnight in a 50°C oven before picking. Thistechnique is extremely effective, staining >95% ofthe ichthyoliths in a sample a pink color (Figure 3).The intensity of the color is dependent on both the

6

PALAEO-ELECTRONICA.ORG

concentration of dye, and the length of time in solu-tion. Time step experiments have shown that theAlizarin protocol does not cause physical degrada-tion of the teeth, even after weeks in the solution,though the color gets progressively darker. How-ever, as the Alizarin staining protocol requires toxicchemicals (KOH) and a second wash, which canincrease the amount of teeth lost to processing, itis generally best saved for particularly challengingresidues, where silica consistently confoundscounts of small ichthyoliths and used consistentlywithin a single record.Heavy liquid separation. Heavy liquids have beenhistorically used to isolate calcium phosphateconodonts from acid-prepared limestone, and themethods have been described extensively else-where (Leiggi and May, 2005). This procedure caneffectively separate ichthyoliths from biogenic silicaand quartz silt, however heavy liquids are expen-sive and often toxic, making them a last resort forichthyolith isolation. We have used both sodiummetatungstate hydrate (Na6W12O39 xH2O) and LSTsolution (heteropolytungstate) as heavy liquids

since both are non-toxic and have low viscosity atroom temperature. Both liquids have the disadvan-tage of being relatively expensive (~$1000/liter)and can be destroyed by contamination with cal-cium. Therefore, the use of heavy liquids on sam-ples containing calcium carbonate grains should beavoided. We use a heavy liquid density of about2.3–2.4 g/cm3 to capture most of the biogenic silicaor 2.85 g/cm3 to separate ichthyoliths from quartzsilt.

In our practice, heavy liquid of suitable densityis poured into a 25–50 mL tube containing the pre-pared sample residues and mixed until the sampleis completely wetted; sufficient heavy liquid shouldbe added to the tube so that the silica can float.The tube is capped and centrifuged for five minutesat 1000–1500 rpm to concentrate the ichthyoliths inthe bottom of the tube. The light fraction is scoopedor poured off the top of the liquid. Both the light andheavy fractions are rinsed in de-ionized water overa 38 μm screen, retaining the rinse solution, thentransferred to filter paper in a funnel and dried in at

FIGURE 3. Paleocene-aged ichthyoliths from ODP Site 1262, stained with Alizarin Red S. The scale bar is 500 μm,with teeth >106 μm in the upper row and teeth <106 μm in the lower. Note that in the coloring effect is present in allteeth, however, the degree of staining varies.

7

SIBERT: QUANTIFYING ICHTHYOLITHS

50°C oven. The dilute heavy liquid is recoveredand cleaned by passing it through a 0.4 µm filter ina vacuum filtration system, and then placed in anoven to evaporate the rinse water and restore itsdensity.

Organic Rich Sediments

While the majority of deep-sea sediments arecarbonate or silica-dominated, there are distincthorizons, such as the Mediterranean sapropels(Cramp and O'Sullivan, 1999), which are organic-rich, and ichthyolith concentration using othermethods is hampered. In addition, modern coralreef sediments, though carbonate dominated, maystill have considerable amounts of organic matter.Organic matter can lead to sediment clumping andadds extra challenges to sample processing. Toaddress this, samples are first disaggregated anddissolved in weak acid, following the carbonatedeep-sea sediments or coral reef sediments proto-cols (see Figure 2), and washed over a 38 μmsieve. However, in the case of high organic mattercontent, there may be numerous organic-richclumps of sediment remaining. Once the samplehas been thoroughly washed, a rinse while on thesieve with dilute (5–10%) bleach solution promotesdisaggregation and dissolution of the remainingorganic matter. However, it is important to note thatbleach and acetic acid produce chlorine gas whenmixed, so caution is advised to ensure that thesample is sufficiently rinsed from acid before anybleach is used.

In the case where disaggregation of organic-rich sediments does not occur with the addition ofde-ionized water or acetic acid, an additional shortsoak in bleach, hydrogen peroxide (H2O2),

BoraxTM, CalgonTM, or OxiCleanTM are a potentialalternatives, although prolonged exposure to anyof these chemicals can damage the ichthyoliths.Most commercial grades of bleach contain perfumeand colorants besides pure sodium hypochlorate,and various formulations produce different results.For instance, in work with Turonian black shalesfrom Ocean Drilling Program Site 1259, weachieved the best disaggregation using pure com-mercial bleach, rather than making a dilute mixture(Bice and Norris, 2005). In our experience, com-mercial grades of bleach vary in their content ofsodium hypochlorate from 5.25% to 6%. Dilutionlowers the pH of bleach solutions, potentiallyincreasing the etching of microfossils with sus-tained contact and may reduce the effectiveness ofthe solution for breaking down organic-rich sedi-ments. However, prolonged exposure to bleach at

any concentration is potentially damaging to theorganic components in ichthyoliths and should belimited if possible. Isolating modern ichthyoliths. Similar to remov-ing organic material from sediments, flesh can beremoved from jaws or skin patches of modernspecimens to isolate taxonomically known fishteeth and shark scales. In this case, the jaw (forfish teeth) or a patch of skin (for shark denticles) isdissected from a modern specimen and placed indilute (5–10%) bleach until all flesh is dissolved,usually 1–4 hours. Since bleach will attack theorganic compounds in teeth and bone as well asthe softer tissues, we recommend removing theichthyoliths from the bleach and washing the newlyisolated modern ichthyoliths as soon as is practi-cal. These isolated modern ichthyoliths are thenwashed over a 38 μm screen and dried in at 50°Coven.

Comments about Ichthyolith-specific Washing and Picking Techniques

Traditional uses of ichthyoliths, for biostratig-raphy or as carriers of isotopes, do not require thatall teeth be retained and accounted for in a sample.However, to assess the ichthyolith accumulationrate, ichthyolith community structure, and the roleof fishes within an ecosystem through time, all ofthe ichthyoliths within a certain size range must bequantified. The methods presented here aim toimprove the fidelity of isolation and concentrationof ichthyoliths, to make this robust quantificationboth possible and repeatable. Due to their smallsize and unusual shape, care must be taken whenhandling the concentrated ichthyolith residue toavoid losing any teeth. As most teeth are triangular,they tend to stick point-down into the sieve whenwashing. Running water up through the back of thesieve, a technique often used when separating bio-logical samples, will help to dislodge any teeth thatare stuck point-down.

Earlier ichthyolith work mounted tooth resi-dues in optical medium and viewed them usingtransmitted light microscopy (Doyle et al., 1977;Doyle and Riedel, 1979a, 1979b, 1985). Transmit-ted light imaging is particularly useful for observingthe details of the interior of the pulp cavity and thestructure of the enamel cap and may have valuefor identification of teeth to taxonomic group (Doyleand Riedel, 1979b; Johns, 1993). Strewn slidesmade by embedding the entire sample residue inCanada Balsam or Norland optical medium canalso be used to count the abundance of extremelysmall teeth, which can be re-located by use of a

8

PALAEO-ELECTRONICA.ORG

England Finder, similar to the study of calcareousnannoplankton. However, there are a number ofdisadvantages of embedding teeth in a mountingmedium, including the formation of bubbles in thepulp cavity, the difficulty in achieving standard ori-entations given the very small size of many teeth,and the three-dimensional aspect of large teeth incontrast with the narrow depth of field in transmit-ted light microscopy. An alternative approach is topick ichthyoliths with a fine paint brush and mountthem with water soluble glue on cardboard micro-paleontology slides. This method retains the mostoptions for quantifying ichthyoliths. It also ensuresthat teeth are not overlooked in original count anal-yses. Once picked, these assemblage slides are aresource, which can be worked with directly or eas-ily be used for many other imaging techniques,including transmitted light microscopy, scanningelectron microscopy, and even microCT or nanoCTscanning. This also leaves the ichthyoliths accessi-ble for geochemical analyses.

BIOLOGICAL ICHTHYOLITH METRICS

Fish Production: Ichthyolith Accumulation Rates

Once isolated and quantified, ichthyoliths pro-vide a unique view of fish production and commu-nity dynamics through time. However, changes insedimentation rate as well as sediment composi-tion and density can have a profound effect on theabsolute abundance of ichthyoliths in a sample,which bias any estimations of fish production orflux. To correct for this, we calculate an ichthyolithaccumulation rate (IAR; eq. 1), yielding a metric ofichthyolith flux of ichthyoliths falling to a fixed areaof seafloor over a fixed time interval. Thus,changes in IAR can be interpreted as increases ordecreases in total ichthyolith production, a proxyfor overall fish production (Sibert et al., 2014, 2016;Sibert and Norris, 2015). IAR in pelagic sedimentsis calculated as:

In the case of sediments from reef matrixcores, which have large fragments of subfossilcoral or mollusk shell (>2 mm), IAR is calculated bynormalizing by the weight of sediments in the sizefraction <2 mm (where the vast majority of teeth

are found) and the number of years represented bya sample. The number of years in a sample can becomputed from U/Th-derived sediment accumula-tion rates (Cramer et al., 2017). This produces anichthyolith abundance accumulation rate (AAR):

This calculation of IAR or AAR normalizes forsedimentation rate and changes in lithology. There-fore, we can compare the flux of ichthyoliths to thesea floor between sites with very different back-ground sedimentation rates, such as between openocean gyre sites and those from the high-produc-tivity equatorial oceans. We can also correct forvariations in sedimentation rate time in a single sitethat result from changes in fish production, sedi-ment delivery, or carbonate dissolution. However,the calculation of accumulation rate is highly sensi-tive to the accuracy of the time scale used to esti-mate sedimentation rate. Bulk density is also acomponent of accumulation rate, but contributesrelatively little to variation in accumulation rate inmost pelagic sediments (see supplement to Sibertet al., 2014). An exception is where there are majorchanges in lithology, such as from carbonates toclaystone or calcareous ooze to limestone; in thesecases accurate measurement of bulk density andsedimentation rate can be important in the calcula-tion of accumulation. Sample size. The size fractions quantified can bestudy-specific, to balance between statistical confi-dence in the data (enough ichthyoliths available),time committed by the researcher, and preserva-tion of other microfossils. We have found throughour work that for pelagic marine carbonates, wheresedimentation rate is 1–2 cm/kyr, quantification ofall ichthyoliths >38 μm in a 10–20cc sample is nec-essary for sufficiently robust abundances of >30–100 teeth/sample. The same sample volume inpelagic red clay can yield hundreds to thousands ofteeth, and statistically significant samples of sev-eral hundred teeth may be found in the >106 μmfraction. In contrast, in coastal sediments and reefcarbonates, the high degree of dilution of ichthyo-liths by other grains and higher sedimentation ratescan require much larger sample volumes to obtainstatistically representative ichthyolith samples. Forexample, in our work in modern Caribbean reefsediments, we routinely sample volumes of 400 cc

Ichthyolith Accumulation Rate = Abundance * Dry Bulk Density * Sedimentation Rates

Ichthyoliths

cm2 kyr ichthyoliths

gram

grams

cm3

cm

kyr =

AARIchthyoliths

gram year

=

equals

ichthyoliths

sample

years

samplegrams

9

SIBERT: QUANTIFYING ICHTHYOLITHS

(about 200 g dry weight) to recover a range of 2 to250 teeth (mean = 74 teeth) and 0–5 denticles persample.

While the abundance of fish may be an indica-tor of primary or export productivity of an ecosys-tem, this is not the only signal recorded in theichthyolith record. The overall efficiency of amarine food web is determined by how many tro-phic steps are needed to transfer the carbon fixedby primary producers up to higher-order consum-ers such as fish. In a large phytoplankton-domi-nated system, such as a modern upwelling zone, amodest primary production will yield abundant fishwith only 1–2 trophic steps. In contrast, a systemdominated by small phytoplankton (such as cyano-bacteria in the open ocean) can require 5–7 trophicsteps to produce a single fish (Moloney and Field,1991; Moloney et al., 1991). While both of theseecosystems may have similar levels of primary pro-duction, the former will produce several orders ofmagnitude more fish biomass than the latter (Iver-son, 1990) and thus should have a significantlyhigher ichthyolith accumulation rate. Indeed, a sub-stantial portion of observed IAR patterns could beaccounted for not by changing net primary produc-tion, but instead by small shifts in the relative abun-dance of certain size classes of phytoplankton.This food web imprint can also be exacerbated bychanges in the efficiency of energy transferbetween trophic levels due to increases ordecreases in metabolic rates of the organisms.Accumulation rates may be also be affected bychanges in habitat: for example, in coral reef sedi-ments, abundances of teeth from coral-associatedtaxa are tightly coupled with reef accretion rates(Cramer et al., 2017).

Ichthyolith accumulation is also influenced bythe production of ichthyoliths by individuals. Spe-cies which put considerable effort into growing theirteeth and have low turnover, or resorb teeth ratherthan shedding them (Bemis et al., 2005), could pro-duce fewer ichthyoliths than a species which pro-duces numerous, but oftentimes less sturdy teethwhich are regularly shed. The majority of the ich-thyolith accumulation rate signal is driven by thesmallest teeth, which likely are derived from a com-bination of small species, juvenile fish, and thepharyngeal tooth battery. At present we are unsureabout the relative contribution of teeth from thesedifferent sources, but we suspect that most teethpreserved in sediments are biased toward thosewith dissolution-resistant enamel caps. The excel-lent preservation of enamel relative to dentine islikely to bias the tooth record toward oral teeth, and

those of species with robust dentitions, and miti-gate against relatively lightly calcified pharyngealteeth and the teeth of some midwater specieswhere an elongate pulp cavity can run almost thefull length of the teeth (Fink, 1981). Indeed, long-term trends in changes in ichthyolith abundance,particularly those associated with shifts in the sizestructure of the assemblage, may reflect an evolu-tionary shift in fish community composition, (e.g.,those documented in Sibert and Norris, 2015 andSibert et al., 2016), rather than a change in overallprimary productivity or food web dynamics.

Fish Community Structure: Ichthyolith Assemblage Metrics

While taxonomic identifications of mostancient ichthyoliths are presently elusive, a consid-erable amount of information about marine verte-brate community composition can be obtained byconsidering the composition of whole ichthyolithassemblages, which preserve snapshots of theentire community, rather than occurrences of a sin-gle species or morphotype. Since ichthyoliths areabundant in most sediment samples, we can eval-uate how the relative abundances of differentmarine vertebrate groups have changed throughtime. Vertebrate community structure. Due to similar-ity in their chemical compositions, both elasmo-branch dermal scales (denticles) and ray-finnedfish teeth are preserved in the ichthyolith record.While the majority of denticles preserved in the ich-thyolith record have been chipped, or preserveonly the enamel crown of the scale, they are dis-tinctive from teeth and readily recognizable as den-ticles (Figure 1). The relative and absoluteabundances of teeth and denticles through timecan be used to study the response of different tro-phic level organisms to global change (Sibert andNorris, 2015; Sibert et al., 2016). Ichthyolith functional group and taxonomiccomposition. Individual ichthyolith size is alsoinformative of evolutionary patterns. While fishtooth size is not necessarily correlated directly withbody size (e.g., deep-sea viperfish of the familyStomiidae have fanglike teeth that are nearly thelength of their head), it is an indicator of diet. Forexample, long, pointed teeth are more likely to beused for handling larger or more active prey. Thesize structure of an ichthyolith assemblage, quanti-fied either through changes in relative abundanceof different size fractions or by measuring thelength of individual teeth (Sibert and Norris, 2015),can reveal evolutionary or ecological trends. For

10

PALAEO-ELECTRONICA.ORG

example, following the Cretaceous-Paleogeneextinction, the maximum size of the largest teeth inan assemblage in open ocean sediments tripledfrom pre-extinction values, suggesting that therewas a radiation of fishes preferring larger prey fol-lowing the extinction event (Sibert and Norris,2015).

While some ichthyoliths are taxonomicallyidentifiable (Figure 4), the majority remain unidenti-fied to taxonomic group at present. Teeth frommodern Caribbean reef sediments have a greatervariety of tooth morphotypes than those frompelagic sediments and can be divided into diet cat-egories such as predators (raptorial or canineteeth), herbivores (incisiform teeth), and duropha-gous invertivores (molariform teeth), producing arecord of fish trophic structure through time (Cra-mer et al., 2017). Utilizing a fish tooth referencecollection for modern Caribbean reef fish (www.ich-thyolith.ucsd.edu), it is also possible to identify sev-eral distinctive tooth types to family level. Pelagicichthyoliths also have discrete morphological char-acters, such as the shape and structure of the pulpcavity, which have been studied in depth for bio-

stratigraphy, and we believe that identification ofeither taxonomic affinity or ecological group willalso become possible for pelagic fishes asresearch progresses. Ichthyolith taphonomy. While ichthyoliths aregenerally resistant to the dissolution effects thatdamage other microfossil groups, there are severaltaphonomic processes that can affect the preser-vation of ichthyoliths. It is rare to find highly frag-mented samples, and ichthyoliths rarely displaypitting or other signs of partial dissolution. How-ever, as many large teeth have an extensive, hol-low pulp cavity, these teeth are prone to splittingdue to mechanical forces, either during preserva-tion or laboratory sediment processing. As thelarge teeth most likely to break are relatively rare ina sample, it is often straightforward to piece a sin-gle large tooth back together following a fracture.We have also occasionally observed iron and man-ganese oxides growing in the pulp cavity of teethwhich can also cause splintering. These sameoxides can also grow around teeth, hiding themfrom observation. Finally, some parts of ichthyolithsare more durable than others. For example, the

FIGURE 4. Examples of select taxonomically identifiable fossil ichthyoliths and modern counterparts. All modern ich-thyoliths were isolated from specimens in the Scripps Marine Vertebrate Collection. The fossil Myctophidae and Triaki-dae specimens are from ODP Site 1262, and are 62 million years old. The Scaridae modern teeth are fromSmithsonian National Museum of Natural History’s Fish Collection and subfossil teeth are from coral reef sedimentcores taken off of the coast of Bocas del Toro, Panama, and are approximately 1200 years old.

11

SIBERT: QUANTIFYING ICHTHYOLITHS

crowns of denticles are much more likely to be pre-served than the subcutaneous base, and the moreheavily enameled tooth tips preserve more often aswell. Despite these potential taphonomic biases,the ichthyolith record is generally well-preserved. Future applications of ichthyoliths. While wehave addressed several applications of the ichthy-olith record here, there are numerous other poten-tial applications. For example, taxonomicallyidentifiable pelagic ichthyoliths can provide signifi-cantly better fossil calibration ages for molecularclock estimates of divergence in open ocean fishlineages, which have a poor body fossil record.Comparison of ichthyolith records with other bio-logical groups present in the same core (e.g., ich-thyoliths and coral community composition in theCaribbean, or fish and foraminifera in the openocean) can reveal trophic or community dynamicsthrough time. IAR or community composition met-rics can also be compared to geochemical proxies,to assess the effects of local or global change onfish population or community ecology. Establishingthe natural abundance, structure, and variability offish communities in coastal, reef, or even lake set-tings, on historic or pre-historic timescales can pro-vide a baseline for separating anthropogenicpressures and climate impacts on economicallysignificant fish stocks. Finally, archaeological mid-dens may have considerable amounts of ichthyo-liths, which could offer insight into how ancienthumans interacted with marine resources. How-ever, archaeological practice commonly studiesonly the >5 mm fraction in screen washings, whichwill fail to recover virtually all teeth (e.g., Powell,2003; Rainsford et al., 2014). Hence, futurearchaeological use of ichthyoliths will require pre-serving at least a known proportion of the finematerial ordinarily discarded during screen wash-ing.

CONCLUSIONS

Ichthyoliths represent an important and under-studied microfossil group that preserves the recordof fishes and sharks at unprecedented temporalresolution. Quantification of the relative and abso-lute abundance of ichthyoliths through time canreveal changing patterns in fish production, foodweb stability, and ecosystem structure throughEarth’s history (including the Anthropocene) andacross global change events throughout much ofthe Phanerozoic. Accurate quantification of thesetrends in ichthyolith accumulation and assemblagestructure relies on quantification of all ichthyolithsin each discrete sample. We have presented a

methodological framework for isolation and quanti-fication of ichthyoliths from most marine sedimenttypes ranging from coral reefs to the open ocean,however, these methods can also be applied tolacustrine or other aquatic deposits. We have fur-ther presented a novel protocol for staining ichthy-oliths pink for easier and more accurate visualidentification using Alizarin Red S. The applicationsof the ichthyolith record include more traditionalbiostratigraphy and geochemistry, alongside fishproduction, evolution, and ancient food web recon-struction. Taxonomic or ecological identification ofichthyoliths will further reveal patterns in fish evolu-tion, shed light on the development and rise todominance of the most diverse group of verte-brates on the planet, and reveal the full magnitudeof change in fish communities resulting from pastand present human activities.

ACKNOWLEDGMENTS

We thank J. Williams, K. McComas, D.Pitassy, H.J. Walker, M. Alvarez, F. Rodriguez, M.P.Concepcion, and A. Castillo for help with develop-ing the modern Caribbean reef fish tooth referencecollection; A. Sanderson and D. Chen for help withisolating and identifying teeth from coral reef cores;B. Oller, C. Carpenter, S. Buckley, and M. Siltanenfor help processing reef sediments; and J. Lyakovfor help with tooth dissolution experiments. E.C.S.was supported on an NSF Graduate Research Fel-lowship. K.L.C. was supported by SmithsonianInstitution MarineGEO Postdoctoral Fellowship andUC San Diego Frontiers of Innovation ScholarsPostdoctoral Fellowship. No permits were requiredfor the described study, which complied with all rel-evant regulations.

REFERENCES

Bellwood, D.R. and Wainwright, P.C. 2002. The historyand biogeography of fishes on coral reefs, p. 5-32. InSale, P.F. (ed.), Coral Reef Fishes: Dynamics andDiversity in a Complex Ecosystem. Academic Press,San Diego.

Bemis, W.E., Giuliano, A., and McGuire, B. 2005. Struc-ture, attachment, replacement and growth of teeth inbluefish, Pomatomus saltatrix (Linnaeus, 1766), ateleost with deeply socketed teeth. Zoology, 108:317-327.

Betancur-R., Broughton, R.E., Wiley, E.O., Carpenter, K.,López, J.A., Li, C., Holcroft, N.I., Arcila, D., San-ciangco, M., and Cureton, J. 2013. The tree of lifeand a new classification of bony fishes. PLoS Cur-rents, 5:1-33.

Bice, K.L. and Norris, R.D. 2005. Data report: Stable iso-tope ratios of foraminifers from ODP Leg 207, Sites

12

PALAEO-ELECTRONICA.ORG

1257, 1258, and 1260 and a cleaning procedure forforaminifers in organic-rich shales. Proceedings ofthe Ocean Drilling Program Scientific Results, 207.Available at http://www-odp.tamu.edu/publications/207_SR/104/104.htm. doi:10.2973/odp.proc.sr.207.104.2005

Broughton, R.E., Betancur, R., Li, C., Arratia, G., andOrtí, G. 2013. Multi-locus phylogenetic analysisreveals the pattern and tempo of bony fish evolution.PLoS Currents, 5, ecurrents.tol.2ca8041495f-fafd0c92756e75247483e.

Cappetta, H. and Schultze, H. 2012. Mesozoic andCenozoic Elasmobranchii: teeth, chondrichthyes,p. 1-512. In Schultze, H. (ed.). Handbook of Palae-oichthyology, 3E. Verlag Publishing, Munich.

Cifelli, R. 1969. Radiation of Cenozoic planktonic fora-minifera. Systematic Biology, 18:154-168.

Cramer, K.L., O'Dea, A., Clark, T.R., Zhao, J., and Nor-ris, R.D. 2017. Prehistorical and historical declines inCaribbean coral reef accretion rates driven by loss ofparrotfish. Nature Communications, 8:14160.doi:10.1038/ncomms14160

Cramp, A. and O'Sullivan, G. 1999. Neogene sapropelsin the Mediterranean: a review. Marine Geology,153:11-28.

Doyle, P., Kennedy, G.G., and Riedel, W.R. 1974. Stratig-raphy. p. 825-905. Thomas, D.A. (ed.), Initial Reportsof the Deep Sea Drilling Project, 26, Durban, SouthAfrica to Fremantle, Australia, September-October1972. National Science Foundation, WashingtonD.C.

Doyle, P.S., Boillot, G., Winterer, E.L., Meyer, A.W.,Applegate, J., Baltuck, M., Bergen, J.A., Comas,M.C., Davies, T.A., Dunham, K.W., Evans, C.A.,Girardeau, J., Goldberg, D., Haggerty, J.A., Jansa,L.F., Johnson, J.A., Kasahara, J., Loreau, J.-P., Luna,E., Moullade, M., Ogg, J.G., Sarti, M., Thurow, J., andWilliamson, M.A. 1988. Remarks on Cretaceous-Ter-tiary ichthyolith stratigraphy in the Atlantic, OceanDrilling Program Leg 103. Proceedings of the OceanDrilling Program, Scientific Results, 103:445-458.

Doyle, P.S., Dunsworth, M.J., and Riedel, W.R. 1977.Reworking of ichthyoliths in eastern tropical Pacificsediments. Deep-Sea Research, 24:181-198.

Doyle, P.S. and Riedel, W.R. 1979a. Cretaceous to Neo-gene ichthyoliths in a giant piston core from the cen-tral North Pacific. Micropaleontology, 25:337-364.

Doyle, P.S. and Riedel, W.R. 1979b. Ichthyoliths: Pres-ent Status of Taxonomy and Stratigraphy of Micro-scopic Fish Skeletal Debris. Scripps Institution ofOceanography Reference Series, 79-16. ScrippsInstitution of Oceanography, University of California,San Diego.

Doyle, P.S. and Riedel, W.R. 1985. Cenozoic and LateCretaceous Ichthyoliths. Plankton Stratigraphy. Cam-bridge University Press, Cambridge.

Edgerton, C.C., Doyle, P.S., and Riedel, W.R. 1977. Ich-thyolith age determinations of otherwise unfossilifer-

ous Deep Sea Drilling Project cores.Micropaleontology, 23:194-205.

Fink, W.L. 1981. Ontogeny and phylogeny of toothattachment modes in actinopterygian fishes. Journalof Morphology, 167:167-184.

Frerichs, W.E. 1971. Evolution of planktonic foraminiferaand paleotemperatures. Journal of Paleontology,45:963-968.

Friedman, M. and Sallan, L.C. 2012. Five hundred mil-lion years of extinction and recovery: a Phanerozoicsurvey of large‐scale diversity patterns in fishes.Palaeontology, 55:707-742.

Gleason, J., Moore, Jr. T., Johnson, T., Rea, D., Owen,R., Blum, J., Pares, J., and Hovan, S. 2004. Age cali-bration of piston core EW9709-07 (equatorial centralPacific) using fish teeth Sr isotope stratigraphy.Palaeogeography, Palaeoclimatology, Palaeoecol-ogy, 212:355-366.

Gleason, J.D., Moore, T.C., Rea, D.K., Johnson, T.M.,Owen, R.M., Blum, J.D., Hovan, S.A., and Jones,C.E. 2002. Ichthyolith strontium isotope stratigraphyof a Neogene red clay sequence: calibrating eoliandust accumulation rates in the central North Pacific.Earth and Planetary Science Letters, 202:625-636.

Gleason, J.D., Thomas, D.J., Moore, T.C., Blum, J.D.,Owen, R.M., and Haley, B.A. 2008. Early to middleEocene Arctic paleoceanography from Nd-Sr isotopestudy of fossil fish debris, Lomonosov Ridge. Geo-chimica et Cosmochimica Acta, 72:A314.

Hallock, P. and Schlager, W. 1986. Nutrient excess andthe demise of coral reefs and carbonate platforms.Palaios, 1:389-398.

Hilgen, F. 1991. Astronomical calibration of Gauss toMatuyama sapropels in the Mediterranean and impli-cation for the geomagnetic polarity time scale. Earthand Planetary Science Letters, 104:226-244.

Hilgen, F.J., Kuiper, K.F., and Lourens, L.J. 2010. Evalu-ation of the astronomical time scale for the Paleo-cene and earliest Eocene. Earth and PlanetaryScience Letters, 300:139-151.

Hull, P.M., Norris, R.D., Bralower, T.J., and Schueth, J.D.2011. A role for chance in marine recovery from theend-Cretaceous extinction. Nature Geoscience,4:856-860.

Ingram, B.L. 1995. High-resolution dating of deep-seaclays using Sr isotopes in fossil fish teeth. Earth andPlanetary Science Letters, 134(3):545-555.

Iverson, R.L. 1990. Control of Marine Fish Production.Limnology and Oceanography, 35:1593-1604.

Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal,K.A., Botsford, L.W., Bourque, B.J., Bradbury, R.H.,Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P.,Kidwell, S., Lange, C.B., Lenihan, H.S., Pandolfi,J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J.,and Warner, R.R. 2001. Historical overfishing and therecent collapse of coastal ecosystems. Science,293(5530):629-638.

Johns, M. 1993. Taxonomy and biostratigraphy of Middleand Upper Triassic ichthyoliths from northeastern

13

SIBERT: QUANTIFYING ICHTHYOLITHS

British Columbia. Unpublished Masters Thesis, Uni-versity of Victoria, Victoria, BC, Canada.

Johns, M.J., Barnes, C.R., and Narayan, Y.R. 2005.Cenozoic and Cretaceous ichthyoliths from theTofino Basin and western Vancouver Island, BritishColumbia, Canada. Palaeontologia Electronica,8.2.29A: 1-202 http://palaeo-electronica.org/paleo/2005_2/icht/issue2_05.htm

Johns, M.J., Barnes, C.R., and Narayan, Y.R. 2006.Cenozoic ichthyolith biostratigraphy; Tofino Basin,British Columbia. Canadian Journal of Earth Sci-ences/ Revue Canadienne des Sciences de la Terre,43(2):177-204.

Kelly, D.C., Bralower, T.J., and Zachos, J.C. 1998. Evolu-tionary consequences of the latest Paleocene ther-mal maximum for tropical planktonic foraminifera.Palaeogeography, Palaeoclimatology, Palaeoecol-ogy, 141(1):139-161.

Leiggi, P. and May, P. 2005. Vertebrate PaleontologicalTechniques (volume 1). Cambridge University Press,Cambridge.

Maisey, J. 1984. Higher elasmobranch phylogeny andbiostratigraphy. Zoological Journal of the LinneanSociety, 82:33-54.

Martin, E. and Haley, B. 2000. Fossil fish teeth as prox-ies for seawater Sr and Nd isotopes. Geochimica etCosmochimica Acta, 64:835-847.

Moloney, C.L. and Field, J.G. 1991. The size-baseddynamics of plankton food webs. I. A simulationmodel of carbon and nitrogen flows. Journal of Plank-ton Research, 13:1003-1038.

Moloney, C.L., Field, J.G., and Lucas, M.I. 1991. Thesize-based dynamics of plankton food webs. II. Simu-lations of three contrasting southern Benguela foodwebs. Journal of Plankton Research, 13:1039-1092.

Near, T.J., Dornburg, A., Eytan, R.I., Keck, B.P., Smith,W.L., Kuhn, K.L., Moore, J.A., Price, S.A., Burbrink,F.T., Friedman, M., and Wainwright, P.C. 2013. Phy-logeny and tempo of diversification in the superradia-tion of spiny-rayed fishes. Proceedings of theNational Academy of Sciences, 110:12738-12743.

Near, T.J., Eytan, R.I., Dornburg, A., Kuhn, K.L., Moore,J.A., Davis, M.P., Wainwright, P.C., Friedman, M.,and Smith, W.L. 2012. Resolution of ray-finned fishphylogeny and timing of diversification. Proceedingsof the National Academy of Sciences, 109:13698-13703.

Nelson, J.S. 2006. Fishes of the World. Wiley, New York.Powell, J. 2003. The fish bone assemblage from the

Cave of Cyclope, Youra: evidence for continuity and

change. British School at Athens Studies, 10:173-179.

Rainsford, C., O'Connor, T., and Miracle, P. 2014. Fish-ing in the Adriatic at the Mesolithic–Neolithic transi-tion: Evidence from Vela Spila, Croatia.Environmental Archaeology, 19(3):311-320.

Randall, J.E. 1961. Overgrazing of algae by herbivorousmarine fishes. Ecology, 42:812.

Scher, H.D. and Martin, E.E. 2004. Circulation in theSouthern Ocean during the Paleogene inferred fromneodymium isotopes. Earth and Planetary ScienceLetters, 228:391-405.

Sibert, E., Norris, R., Cuevas, J., and Graves, L. 2016.Eighty-five million years of Pacific Ocean gyre eco-system structure: long-term stability marked by punc-tuated change. Proceedings of the Royal Society ofLondon B: Biological Sciences, 283(1831).doi:10.1098/rspb.2016.0189

Sibert, E.C., Hull, P.M., and Norris, R.D. 2014. Resil-ience of Pacific pelagic fish across the Cretaceous/Palaeogene mass extinction. Nature Geoscience,7:667-670.

Sibert, E.C. and Norris, R.D. 2015. New Age of Fishesinitiated by the Cretaceous−Paleogene mass extinc-tion. Proceedings of the National Academy of Sci-ences, 112:8537-8542.

Smit, J. 1982. Extinction and evolution of planktonic fora-minifera after a major impact at the Cretaceous/Ter-tiary boundary. Geological Society of AmericaSpecial Papers, 190:329-352.

Song, J. and Parenti, L.R. 1995. Clearing and stainingwhole fish specimens for simultaneous demonstra-tion of bone, cartilage, and nerves. Copeia, 1:114-118.

Sprules, W.G., and Munawar, M. 1986. Plankton SizeSpectra in Relation to Ecosystem Productivity, Size,and Perturbation. Canadian Journal of Fisheries andAquatic Sciences, 43(9):1789-1794.

Taylor, W.R. 1967. An enzyme method of clearing andstaining small vertebrates. Proceedings of the U.S.National Museum, 122(3596):1-17.

Turner, S. 2004. Early vertebrates: analysis from micro-fossil evidence, p. 67-94. In Arratia, G., Wilson,M.V.H., and Cloutier, R. (eds.), Recent Advances inthe Origin and Early Radiation of Vertebrates, 65.Verlag Dr. Friedrich Pfeil, Munich.

Westerhold, T., Roehl, U., Raffi, I., Fornaciari, E.,Monechi, S., Reale, V., Bowles, J., and Evans, H.F.2008. Astronomical calibration of the Paleocenetime. Palaeogeography, Palaeoclimatology, Palaeo-ecology, 257(4):377-403.

14