shell hardness and compressive strength of the eastern ... · c. virginica than for c. ariakensis...

Shell Hardness and Compressive Strength of theEastern Oyster, Crassostrea virginica, and the Asian

Oyster, Crassostrea ariakensis

SARA A. LOMBARDI1,*,†, GRACE D. CHON1,¶, JAMES JIN-WU LEE2,§,HILLARY A. LANE1, AND KENNEDY T. PAYNTER1,3

1Department of Biology, University of Maryland, College Park, College Park, Maryland 20742;2Ceramics Division, National Institute of Standards and Technology, Gaithersburg, Maryland 20899;

and 3Chesapeake Biological Laboratory, University of Maryland Center for Environmental Science, P.O.Box 38, 1 Williams Street, Solomons, Maryland 20688

Abstract. The valves of oysters act as a physical barrierbetween tissues and the external environment, thereby pro-tecting the oyster from environmental stress and predation.To better understand differences in shell properties andpredation susceptibilities of two physiologically and mor-phologically similar oysters, Crassostrea virginica andCrassostrea ariakensis, we quantified and compared twomechanical properties of shells: hardness (resistance to ir-reversible deformation; GPa) and compressive strength(force necessary to produce a crack; N). We found nodifferences in the hardness values between foliated layers(innermost and outermost foliated layers), age class (C.virginica: 1, 4, 6, 9 years; C. ariakensis: 4, 6 years), orspecies. This suggests that the foliated layers have similarproperties and are likely composed of the same material.The compressive force required to break wet and dry shellswas also not different. However, the shells of both six- andnine-year-old C. virginica withstood higher compressiveforce than C. virginica shells aged either one or four, andthe shells of C. ariakensis at both ages studied (4- and6-years-old). Differences in ability to withstand compres-

sive force are likely explained by differences in thicknessand density between age classes and species. Further, wecompared the compressive strength of differing ages ofthese two species to the crushing force of common oysterpredators in the Chesapeake Bay. By studying the physicalproperties of shells, this work may contribute to a betterunderstanding of the mechanical defenses of oysters as wellas of their predation vulnerabilities.

Introduction

Adult oysters possess two calcareous valves connected bya hinge ligament that acts as a spring and allows an oysterto separate the valves (gape), thereby permitting an ex-change between the internal and external environments(Carriker, 1996). Each valve is composed of three primarylayers. The outermost layer is the periostracum, which pri-marily consists of an organic (conchiolin) matrix and pro-tects the shell from corrosion (Galtsoff, 1954, 1964; Bottjerand Carter, 1980; Carriker, 1996; Fig. 1). The middle layeris the prismatic layer, which contains curved, wedge-shapedprisms of calcite crystals within a conchiolin matrix (Galt-soff, 1954, 1964; Mount et al., 2004). The innermost layer,nearest the pallial space, is the calcite-ostracom layer, alsoknown as the foliated layer. The foliated layer constitutesthe majority of the shell and comprises sheets of calcite thatare associated with providing shell strength (Galtsoff, 1954,1964; Chateigner et al., 2002; Mount et al., 2004). Since thefoliated layer is composed of parallel sheets of calcitecrystals, its microscopic structure is similar to that of thenacre layer of aragonite structures (Checa et al., 2007).

Received 29 March 2013; accepted 26 November 2013.* To whom correspondence should be addressed. E-mail: SaraAnn

[email protected]† Current address: Department of Environmental Science, American

University, 4400 Massachusetts Ave. NW, Washington, DC 20016.¶ Current address: Department of Natural Sciences, Hawaii Pacific Uni-

versity, 45-045 Kamehameha Highway, Kaneohe, HI 96744.§ Current address: Department of Biological Sciences, Marshall Univer-

sity, 1 John Marshall Drive, Science Building Room 350, Huntington, WV25755.

Reference: Biol. Bull. 225: 175–183. (December 2013)© 2013 Marine Biological Laboratory

175

Within oyster shells, the foliated layer acts as the maindefense that predators must penetrate to access oyster tissue.Overall, the shells assist in survival, as they provide spacefor internal organs and shield tissues from outside forcesthat include predators as well as environmental stressorssuch as salinity changes, desiccation, and anoxic or contam-inated water (Carriker, 1996; White and Wilson, 1996).Therefore, understanding the properties of the shells, inparticular the foliated layer, can contribute to understandingoyster predation defenses and survival. Ultimately, the me-chanical defense provided by the shells enables oysters toperform essential ecosystem services such as habitat forma-tion, water filtration, and benthic-pelagic coupling in marineand estuarine systems (see Newell, 1988; Meyer andTownsend, 2000; Steimle and Zetlin, 2000; Porter et al.,2004; Grabowski et al., 2005; see Geraldi et al., 2009 for areview of oyster ecosystem services).

Throughout the past century the Chesapeake Bay popu-lation of the Eastern oyster, Crassostrea virginica (Gmelin,1791), has been in decline due to over-harvesting, habitatdestruction, and the introduction and spread of parasiticdiseases such as Dermo, caused by Perkinsus marinus(Mackin, Owen, and Collier) Levine, 1978 (Newell, 1988;Rothschild et al., 1994; Ford and Tripp, 1996). However,the morphologically and taxonomically similar Asian oys-

ter, Crassostrea ariakensis (Fujita, 1913), is more resistantto P. marinus infection and more resilient to its effects wheninfected (Calvo et al., 2001; NRC, 2004). This led to theconsideration of C. ariakensis for introduction to the Ches-apeake Bay to augment the native population, thereby aid-ing in ecosystem services and providing an alternative spe-cies for the commercial oyster industry. The potentialintroduction of C. ariakensis to part of the native range of C.virginica resulted in numerous studies comparing the anat-omy and physiology of these two species (e.g., Calvo et al.,2001; Matsche and Barker, 2006; Harding, 2007; Harlan,2007; Lombardi, 2012; Lombardi et al., 2013). During onesuch study, it was observed that when notching shells witha grinding wheel, considerable more effort was required forC. virginica than for C. ariakensis (Lombardi and Paynter,unpubl.). This observation illustrated the need for a morein-depth investigation of the mechanical properties of thevalves of these oysters, in order to understand the mecha-nisms for these apparent differences in strength and dura-bility that may translate to differences in survival.

In this study, we investigate factors potentially affectingoyster shell strength, including age, species, and shell wet-ness, and we discuss the implications of these factors forpredation pressures. The mechanical properties of hardness(resistance to irreversible deformation) and compressivestrength (force necessary to produce a crack in a material) ofoyster shells were measured in various age classes of C.ariakensis and C. virginica. Since oyster shell layers accu-mulate as the animal grows, different layers may havedifferent properties. Therefore, shells were also tested fordifferences in the hardness between superficial and deepfoliated layers. Further, the compressive load at fracture, ametric of the ability to withstand pressure without cracking,was measured on the whole shell.

On the basis of perceived shell durability, we hypothe-sized that C. virginica shells would be more resistant tofracture and indentation, and thus harder and able to with-stand a higher compressive force than C. ariakensis shells.We further hypothesized that age would be correlated withthickness and that older shells would be able to withstand ahigher compressive force without cracking than youngeroyster shells. The results of this study provide insight intothe shell properties of two similar oyster species and maycontribute to an understanding of mechanisms for defenseagainst predators and for survival.

Materials and Methods

Oyster preparation

Crassostrea virginica individuals (ages 1, 4, 6, and 9years) were obtained from hatchery-planted bars in theChoptank River in the northern Chesapeake Bay, and Cras-sostrea ariakensis individuals (ages 4 and 6 years) wereobtained from University of Maryland Center for Environ-

Figure 1. Primary layers of Crassostrea virginica oyster shells. Thiscross-section of a valve illustrates the primary layers of the oyster shellfrom superficial to deep: periostracum (A), prismatic layer (B), and thecalcite-ostracom, or foliated, layer (C). Each of these layers is present inboth the left and right valves. Modified from Galtsoff (1964).

176 S. A. LOMBARDI ET AL.

mental Science’s Horn Point Laboratory Oyster Hatcheryprogram. Each oyster was scrubbed with a wire brush toremove any debris and epifauna and then shucked to isolateshells from oyster tissue. Shell height (mm) and total mass(g) of each individual were also measured.

Hardness

The hardness, measured in gigapascals (GPa), of four C.virginica shells from each age class (1, 4, 6, and 9 years)and four C. ariakensis shells from each age class (4 and 6years) was tested. A diamond wheel was used to cut twopieces (about 10 mm by 20 mm) from the flattest part ofeach valve. One piece was used to test hardness at thedeepest foliated layer, and the second piece was used to testhardness at the most superficial foliated layer. In all sam-ples, the area where the adductor muscle attached to thevalve was not used so as to maintain consistent valvesampling.

Each valve sample was embedded in resin and polishedwith a Buehler EcoMet 4000 Variable Speed Grinder-Pol-isher (Lake Bluff, IL) with water or suspensions of abrasiveparticles between the samples and grinding pads, in order tofacilitate the collection of clear microindentations (Chinn,2002). Each sample was polished for 10-40 min at each gritsize (400, 600, and 1200 grit with water, microcloth withBuehler TexMet 1.0-�m suspension, and microcloth with0.5-�m diamond suspension). Between each grit size, allsamples were placed in a sonicator for 5 min and thenobserved under a microscope to inspect for uniform surface

scratches to determine if the sample was ready to progressto the next grit size. After the finest grit polishing, thesamples were mirror-smooth (Fig. 2A).

Each valve sample was taken to the National Institute ofStandards and Technology where five indents on each sam-ple were made using a Vickers pyramidal diamond micro-indenter (Zwick, GA; Fig. 2B). Each indent image (Fig. 2C)was captured using a Leco Olympus PMG3 microscope(Olympus, PA). Hardness values were calculated using thestandard Vickers hardness (HV) equation:

HV � 1.854 � F/d2,

where F is the applied force (1 kg and 9.8 m/s2) and d is themean length (mm) of the indentation diagonals. Mean hard-ness values for each sample were calculated.

Analysis of hardness between superficial and deep foli-ated layers, species, and age (4- and 6-year-olds) was per-formed using a fixed-factor split-plot analysis of variance(ANOVA), with oyster age as the main plot and eachfoliated layer as the subplot. A separate one-way fixed-factor ANOVA was conducted to test for intraspecific hard-ness differences among dry specimens of C. virginica atdifferent ages (1, 4, 6, and 9 years) using the foliated layerwith the lowest hardness value.

Compressive strength

Intraspecific and interspecific comparisons. Interspecificanalyses based on age classes (4 and 6 years) were con-

Figure 2. Protocol necessary to prepare oyster samples for microindentations. After samples of oyster shellswere cut, they were placed in resin and polished to create a reflective surface (Panel A). A microindenter (PanelB) placed a diamond tip into a sample to create an indentation in the shell (Panel C). The square in Panel C isa microindentation into a Crassostrea ariakensis shell sample.

177CRASSOSTREA SHELL MECHANICS

ducted using 10 dry shells of each species. Additionally, 10dry C. virginica shells aged one and nine were used forintraspecific analyses. Before the compressive strength wasmeasured, shell thickness within 5 mm of the ventral shellmargin of the curved valve was measured using digitalcalipers (mm). The compressive force, measured in newtons(N), needed to fracture each shell was determined by usingan Instron load compression machine with a flat load cell(Instron, PA). The left and right valves of each shell wererealigned to mimic a closed live oyster. The flat right valvewas placed on the base of the machine, while the cupped leftvalve pointed upward with the tallest point of the shellaligned with the center of the machine. Compressivestrength was assessed by increasing the force until the shellcracked and there was a major drop in load compressionforce. After the drop in force was observed, the oyster wasexamined to confirm the presence of a crack. In all samples,the crack was visually identified and went through all layersof the shell. Typically both valves were cracked, though onoccasion only the upper, cupped valve cracked. We ob-served complete radial cracks that originated in the center ofthe shell and spread outward to the edge.

A two-way fixed-factor ANOVA was used to test fordifferences in compressive strength based on age and spe-cies and to test for any interactions between the two. Ad-ditionally, a one-way fixed-factor ANOVA was used to testfor differences in compressive strength between all ages (1,4, 6, 9 years) of C. virginica. To determine if thicknesscould explain the patterns observed, we tested whetherthickness differed between age classes or species byusing a two-way fixed-factor ANOVA and two separateanalysis of covariance (ANCOVA) models (one with spe-cies as a factor and a second with age class) with thicknessas a covariate.

Wet versus dry shells. The relationship between shell wet-ness and compressive strength was also investigated usingsix-year-old oyster shells; five shells of each species werekept wet, and five shells of each species were placed in a60 °C oven for 72 h to dry. Dry shells were stored in adesiccator at room temperature (about 24 °C) until com-pressive strength data were collected using the proceduredescribed in the previous section on Intraspecific and inter-specific comparisons. A separate two-way fixed-factorANOVA was used to test for significant differences incompressive strength between wet and dry shells, species,and any interaction between the two factors.

Shell density

Density was measured in 10 C. virginica shells for eachage (1, 4, 6, and 9 years) and 10 C. ariakensis shells for eachage (4 and 6 years). Shells measured for density weredifferent shells than those used for either compressive

strength or hardness tests but were from the same sourceand were conditioned identically. The volume of each pairof valves was calculated by displacement using graduatedcylinders, and the dry mass of the shells weighed on amicrobalance. Using these values, density (g/ml) was cal-culated for each shell. A separate two-way fixed-factorANOVA with age and species as factors was used to com-pare shell densities.

For all comparisons, statistical analyses were performedin JMP (JMP, NC) and R 2.11.1 (R Core Team, 2013). Insome cases, log-transformed data were used to meet nor-mality or homogeneity of variance assumptions. Since the Pvalues between transformed and non-transformed data werecomparable and the accept/reject decisions were the same,results are presented on non-transformed data to ease inter-pretation.

Results

Hardness

Intraspecific Crassostrea virginica comparisons (ages 1, 4,6, 9 years). The mean (�SEM) shell hardness values forone-year-old individuals of C. virginica were 1.17 � 0.17GPa, 1.55 � 0.22 GPa for four-year-olds, 1.47 � 0.16 GPafor six-year-olds, and 1.20 � 0.14 GPa for nine-year-olds.Hardness values were not significantly different betweenage classes of C. virginica (P � 0.368). As there was nosignificant difference in hardness between outer and innerfoliated layers (P � 0.274, see the following Results sectionon interspecific comparisons), the intraspecific analysis wasperformed using the lower hardness value from each shell.This was done because most predators would need to pen-etrate only one valve, and therefore we chose to use theweaker valve.

Interspecific comparison (ages 4 and 6 years). When weperformed the split-plot analysis, no significant two-way orthree-way interactions were found for shell hardness be-tween age, foliated layer, or species (in all comparisons P �0.05). Crassostrea ariakensis exhibited a shell hardness of1.45 � 0.13 GPa, which was not significantly different fromthat of C. virginica shells (1.77 � 0.13 GPa; P � 0.090).Shell hardness values did not significantly vary betweenfour-year-old (1.64 � 0.15 GPa) and six-year-old oysters(1.57 � 0.12 GPa; P � 0.635) or between foliated layers(�deep � 1.50 � 0.12 GPa; �superficial � 1.72 � 0.14 GPa;P � 0.274). It should be noted that C. virginica shell valuesdiffer from those in the intraspecific analysis (previousResults section), because the intraspecific analysis usedonly one foliated layer per oyster, whereas this analysis usedtwo samples per shell.


Compressive strength

Intraspecific Crassostrea virginica comparisons. No signif-icant difference in compressive strength was found betweenone- (853.9 � 169.1 N) and four-year-old C. virginicashells (1089.8 � 90.8 N; P � 0.235). However, a significantincrease in compressive force was needed to crack both six-and nine-year-old shells (mean forces were 3896.1 �

560.7 N and 3513.1 � 476.3 N, respectively) compared toone- and four-year-old shells (mean forces were 853.9 �169.1 N and 1089.8 � 90.8 N; P � 0.001, P � 0.001,respectively).

Interspecific analysis (ages 4 and 6 years)

A significant interaction was found between age andspecies for compressive strength (P � 0.001; Fig. 3A). The

Figure 3. Compressive force, thickness, and density of Crassostrea virginica and Crassostrea ariakensisshells. Panel A: Compressive force of different ages of C. ariakensis and C. virginica. Significantly more force(N) was necessary to produce a crack in six- and nine-year-old C. virginica shells than in shells of four-year-oldC. virginica or four- and six-year-old C. ariakensis shells. Panel B: Thickness by species for each age class.Shells of age six C. virginica were significantly thicker than shells of age four C. virginica and four and six C.ariakensis. Panel C: Shell density by species. Crassostrea virginica shells were more dense than C. ariakensisshells. In all panels, C. virginica values are shown in a darker color, error bars depict standard error, and differentletters indicate significantly different values at an alpha of 0.05.


ability to withstand compressive force did not statisticallyvary between four- (971.4 � 84.1 N) and six- (1522.0 �315.1 N) year-old C. ariakensis (P � 0.109). However,six-year-old C. virginica shells were able to withstand asignificantly higher compressive force (3896.1 � 560.7 N)than both four- (P � 0.001) and six- (P � 0.002) year-oldC. ariakensis shells. Compressive strength did not differbetween shells of four-year-old C. virginica or age four- orsix-year-old C. ariakensis (P � 0.538).

Thickness, density, and wetness. A significant exponentialrelationship between the compressive force required to pro-duce a crack (y) and shell thickness (x) was found (y �523.6e0.381x, R2 � 0.4657; P � 0.001; Fig. 4)—thickeroysters withstood higher compressive force. ANOVA re-sults indicated that thickness varied on the basis of speciesand age class (significant interaction between species andage, P � 0.003). At age four, the thickness of C. virginicashells (2.47 � 0.26 mm) was not significantly different fromthat of C. ariakensis (2.07 � 0.24 mm; P � 0.275); how-ever, at age six, C. virginica shells were more than twice asthick as C. ariakensis shells (�C. ariakensis � 2.01 � 0.38mm; �C. virginica � 4.53 � 0.42 mm; P � 0.001). ANCOVAindicated no significant interaction between thickness andage (P � 0.172) or thickness and species (P � 0.871) andthat in each case thickness had a significant effect (P �0.001; P � 0.002, respectively). After accounting for thick-ness using ANCOVA, species were no longer different (P �0.358), but a significant trend persisted with age—valves ofolder oysters had a higher compressive strength thanyounger valves (P � 0.004).

When analyzing density, no significant interaction be-tween species and age was found (P � 0.923). Age also hadno effect on density (P � 0.876), but C. virginica shellsexhibited a mean density of 2.18 � 0.06 g/ml, which wasmore dense than C. ariakensis shells (1.44 � 0.12 g/mL;Fig. 3C; P � 0.001).

No interaction was found between wetness and species(P � 0.930), and the force needed to produce a crack did notdiffer between wet (2512.8 � 441.8 N) and dry (2191.5 �532.7 N) shells (P � 0.547).

Discussion

This study investigated the mechanical properties of com-pressive strength and hardness across different ages of adultCrassostrea virginica and C. ariakensis shells. We found nodifferences in shell hardness by age, layer (innermost oroutermost foliated layer), or species. In view of the hardnessdata, we believe the foliated layers of adult C. virginica andC. ariakensis likely have similar compositions, giving riseto similar strength and defenses. As similar calcite config-uration likely yields similar strength, our findings are inagreement with those of Checa et al. (2007), who found thatthe crystallographic structure of the foliated layer had aconsistent arrangement across bivalve species. Further,compressive force data revealed that the compressionstrength of C. virginica shells increased with age, but nodifference was found in compression strength between C.ariakensis age classes. Interspecifically, ages six and nine ofC. virginica could withstand significantly more compres-sion force than either age class (4 or 6 years) of C. ariak-ensis.

In particular, the compression strength of six-year-old C.ariakensis shells was 61% lower than that of six-year-old C.virginica shells; this difference is similar to the one (64%)observed between spat (oysters less than a year old) of thesespecies (Newell et al., 2007). Interestingly, however, at agefour no significant difference (�10%) existed between spe-cies. Therefore, the ratio of compressive strength betweenC. virginica and C. ariakensis individuals likely varies byage. While the reason for this is unknown, it may reflectdifferences in growth pattern, growth rate, energy alloca-tion, or life-history strategies between these species.

Figure 4. Thickness and compressive force. Prior to testing the compressive force required to fracture theshell, the thickness within 5 mm of the ventral shell margin of the curved valve was measured using digitalcalipers. This figure illustrates the relationship between thickness (x) and the compressive force required toproduce a crack (y) in four- and six-year-old specimens of Crassostrea virginica and Crassostrea ariakensis.


Compressive strength was not different between speciesat age four years, but six-year-old C. virginica shells with-stood a higher compressive force than six-year-old C. ari-akensis, a pattern that may be explained by differences indensity and thickness. The shells of C. ariakensis were 34%less dense than those of C. virginica. Regression analysisindicated a significant relationship between valve thicknessand compressive strength, and six-year-old C. virginica hadthe thickest valves. Additionally, when an ANCOVA wasperformed with thickness as a covariate, species no longerhad an effect on compression strength. Local curvaturesnotwithstanding, the dependance of loads required for frac-tures originating in the interior section of the shells andother material are dependent on the square of thickness, asseen in work presented in Qasim et al. (2006, 2007). There-fore, the protective nature of shell thickness versus the loadrequired to fracture the shell under a point load reveals thatsmall attenuations in shell thickness impact fracture loadsby a squared relation. However, our data do not fully adhereto this pattern. We believe the discrepancy is due to theshape of the shells, as curvature can distribute the load overthe entire structure, thereby influencing the compressivestrength (Sayegh and Dong, 1970). Therefore, the mechan-ical defenses of shells likely include the foliated layerstructure and the curve and thickness of the shell. In sum-mary, we believe that the compressive strength differencesbetween species can be primarily explained by differencesin shell thicknesses and density; however, those factors donot explain ontogenetic differences based on age.

Since many predators penetrate through the shell to ac-cess internal oyster tissue (White and Wilson, 1996), iden-tifying differences in shell strength by age and species couldhave implications for restoration and management of thenative population and the stability of the fishing industry.Therefore, the differences in strength, illustrated by theability to withstand compressive force, between C. ariak-ensis and C. virginica could translate into differences inpredation. Although predators vary by location, commonpredators of oysters include fish (namely drums, sheeps-heads, rays, and skates); crabs; snails (especially drills); seastars; and flatworms (Mackin, 1959). However, specieshave different mechanical abilities to penetrate the shell. Forinstance, the cownose ray, Rhinoptera bonasus (Mitchill,1815), is capable of bite forces of about 220 N (Maschner,2000); and the bite of the large black drum, Pogoniascromis (Linnaeus, 1766), can reach up to 1250 N (Grubich,2005). Further, the rock crab, Cancer irroratus Say, 1817,has demonstrated chelae crushing forces of up to 75 N(Block and Rebach, 1998); and the blue crab, Callinectessapidus Rathbun, 1896, has exhibited a mean chelae forceof 27 N (Singh et al., 2000). In contrast, the mean compres-sive strength of C. virginica spat ranged from 51 to 71 Nand that of C. ariakensis from 18 to 25 N, depending on thepresence of predators (Newell et al., 2007); in this study we

found the compressive force needed to crack the shells ofone- and four-year-old C. virginica and four-year-old C.ariakensis was about 1000 N, whereas six- and nine-year-old C. virginica withstood more than 3500 N. Thus, if apredator attacks the center of the shell, then both spat andadult C. virginica and C. ariakensis should be able towithstand cownose ray, rock crab, and blue crab predation,but C. ariakensis and young (ages four and younger) C.virginica are still vulnerable to drum predation.

Another factor that may influence shell properties is theratio of chalky deposits to foliated sheets within the valves.Naturally interspersed between foliated sheets are chalkydeposits (Galtsoff, 1964; Carriker, 1996). Although bothchalky and non-chalky layers are made of similar chemicalelements, they have different chemical compositions, andchalky material is more porous than the folia (Yoon et al.,2009; Lee et al., 2011). Galtsoff (1964) found that 50% ofC. virginica samples had no chalky patches, and only 3%had more than 75% of the shell composed of such deposits.However, no studies have quantified the chalky proportionsin C. ariakensis. In this study we were interested in assess-ing the hardness of the foliated layer; however, while col-lecting data we encountered regions of high variabilitywithin C. ariakensis. We believe these represented samplestaken in the chalky layer, and we excluded them fromanalysis (Appendix Fig. 1). Although some discrepancyexists in the literature (Lee et al., 2011), Yoon et al. (2009)and Ullmann et al. (2010) found the chalky layer of Cras-sostrea oysters species to be less dense than the foliatedlayers. Therefore, since regions of high variability werefound only in C. ariakensis valves that were also less densethan those of C. virginica, we hypothesize that C. ariakensismay have a higher proportion of chalky material than C.virginica. A difference in this proportion could also explainthe differences in strength and durability observed at themacroscopic level—that is, the greater difficulty that Lom-bardi and Paynter (unpubl. data) experienced in grinding C.virginica shells compared to C. ariakensis shells. Despiteproviding less strength, a higher proportion of chalky ma-terial may be advantageous in an area dominated by gape-or claw-limited predators because chalk deposits may in-crease shell thickness at minimal cost to the oyster. Thus,different proportions of chalky material to folia likely in-fluence the mechanical defenses the shell provides and mayindicate different predation styles and pressures in nativeranges. Overall, the shells of C. ariakensis may be lessdurable and strong due to a greater proportion of chalkymaterial, though further studies are needed to test this hy-pothesis.

In summary, we found that the shells of six- and nine-year-old specimens of C. virginica withstood significantlyhigher compression force than any age class of C. ariakensisstudied (4- and 6-year-old). The higher compressivestrength observed for C. virginica shells at age six is likely


due to thicker valves and denser shells compared to C.ariakensis. We found no differences in hardness valuesbetween any age class, foliated layer, or species; however,differences in density and observed variations in hardnesswithin C. ariakensis (Appendix Fig. 1) may indicate differ-ent shell compositions between species. The findings of thisresearch can be applied to oyster management decisions, aswell as to oyster-inspired bioceramic technology. Moreover,this work contributes to an understanding of shell mechan-ics and sheds light on similarities and differences betweentwo closely related species while identifying potential dif-ferences in predation vulnerability.

Acknowledgments

Support for this research was provided by grants to theUniversity of Maryland from the Oyster Recovery Partner-ship and the Howard Hughes Medical Institute Undergrad-uate Science Education Program. This project would nothave been possible without the encouragement, assistance,and expertise of Peter Lucas, Erin Vogel, and ShannonMcFarlin from George Washington University. We thankBrian Lawn from the National Institute of Standards andTechnology for his materials science expertise and guid-ance. Further, we are grateful to Adriane Michaelis, KarenKesler, Rebecca Kulp, Emma Weaver, and Drew Needhamfor their assistance with shell polishing and data collection,and to Robert Bonenberger, Vincent Politano, Jeffrey Jen-sen, and William Higgins for their academic and logisticalhelp. Additionally, we thank the Horn Point Oyster Hatch-ery program for supplying the Crassostrea ariakensis usedin this study. We also thank the editor and two anonymousreviewers for constructive comments that helped strengthenthe manuscript.

Literature Cited

Block, J. D., and S. Rebach. 1998. Correlates of claw strength in therock crab, Cancer irroratus (Decapoda, Brachyura). Crustaceana 71:468–473.

Bottjer, J. F., and J. G. Carter. 1980. Functional and phylogeneticsignificance of projecting periostracal structure in the Bivalvia (Mol-lusca). J. Paleontol. 54: 200–216.

Calvo, G. W., M. W. Luckenback, S. K. Allen, Jr., and E. M. Burreson.2001. A comparative field study of Crassostrea ariakensis (Fujita1913) and Crassostrea virginica (Gmelin 1791) in relation to salinity inVirginia. J. Shellfish Res. 20: 221–229.

Carriker, M. R. 1996. The shell and ligament. Pp. 75–168 in TheEastern Oyster, Crassostrea virginica, V. S. Kennedy, R. I. E. Newell,and A. F. Eble, eds. Maryland Sea Grant Publication, College Park,MD.

Chateigner, D., M. Morales, and F. M. Harper. 2002. QTA of pris-matic calcite layers of some bivalves, a link to trichite ancestrals.Mater. Sci. Forum 408: 1687–1692.

Checa, A. G., F. J. Esteban-Delgado, and A. B. Rodrıguez-Navarro.2007. Crystallographic structure of the foliated calcite of bivalves. J.Struct. Biol. 157: 393–402.

Chinn, R. E. 2002. Ceramography: Preparation and Analysis of Ce-

ramic Microstructures. ASM International, Materials Park, OH. Pp.35–43.

Ford, S. E., and M. R. Tripp. 1996. Diseases and defense mechanisms.Pp. 581–660 in The Eastern Oyster, Crassostrea virginica, V. S.Kennedy, R. I. E. Newell, and A. F. Eble, eds. Maryland Sea GrantPublication, College Park, MD.

Galtsoff, P. S. 1954. Recent advances in the studies of the structure andformation of the shell of Crassostrea virginica. Proc. Natl. ShellfishAssoc. 45: 116–135.

Galtsoff, P. S. 1964. The American oyster: Crassostrea virginica (Gme-lin). Fish. Bull. 64: 1–480.

Geraldi, N. R., S. P. Powers, K. L. Heck, and J. Cebrian. 2009. Canhabitat restoration be redundant? Response of mobile fishes and crus-traceans to oyster reef restoration in marsh tidal creeks. Mar. Ecol.Prog. Ser. 389: 171–180.

Grabowski, J. H., A. R. Hughes, D. L. Kimbro, and M. A. Dolan. 2005.How habitat setting influences restored oyster reef communities. Ecol-ogy 86: 1926–1935.

Grubich, J. R. 2005. Disparity between the feeding performance andpredicted muscle strength in the pharyngeal musculature of black drum,Pogonias cromis (Sciaenidae). Environ. Biol. Fishes 74: 261–272.

Harding, J. M. 2007. Comparison of growth rates between diploidDEBY eastern oysters (Crassostrea virginica, Gmelin 1971), triploideastern oysters, and triploid suminoe oysters (C. ariakensis, Fujita1913). J. Shellfish Res. 26: 961–972.

Harlan, N. P. 2007. A comparison of the physiology and biochemistry ofthe eastern oyster, Crassostrea virginica and the Asian oyster, Cras-sostrea ariakensis. Master’s thesis. University of Maryland, CollegePark, MD.

Lee, S.-W., Y.-N. Jang, K.-W. Ryu, S.-C. Chae, Y.-H. Lee, and C.-W.Jeon. 2011. Mechanical characteristics and morphological effect ofcomplex cross structure on biomaterials: fracture mechanics and mi-crostructure of chalky layer in oyster shell. Micron 42: 60–70.

Lombardi, S. A. 2012. Comparative physiological ecology of the easternoyster, Crassostrea virginica, and the Asian oyster, Crassostrea ari-akensis: an investigation into aerobic metabolism and hypoxic adapta-tions. Ph.D. dissertation, University of Maryland, College Park, MD.

Lombardi, S. A., N. P Harlan, and K. T. Paynter. 2013. Survival,acid-base balance and gaping responses of the Asian oyster Cras-sostrea ariakensis and the eastern oyster Crassostrea virginica duringclamped emersion and hypoxic immersion. J. Shellfish Res. 32: 409–415.

Mackin, J. G. 1959. Mortalities of oysters. Proc. Natl. Shellfish Assoc.50: 21–40.

Maschner, R. P., Jr. 2000. Studies of the tooth strength of the Atlanticcow-nose ray, Rhinoptera bonasus. Master’s thesis, California StatePolytechnic University.

Matsche, M., and L. Barker. 2006. Juvenile oyster mortality followingexperimental exposure to anoxia/hypoxia—C. ariakensis vs. C. virgi-nica. Report to the Maryland Department of Natural Resources. Avail-able from the Maryland Department of Natural Resources, Annapolis.

Meyer, D. L., and E. C. Townsend. 2000. Faunal utilization of createdintertidal eastern oyster (Crassostrea virginica) reefs in the southeast-ern United States. Estuaries 23: 34–45.

Mount, A. S., A. P. Wheeler, R. P. Paradkar, and D. Snider. 2004.Hemocyte-mediated shell mineralization in the eastern oyster. Science304: 297–300.

Newell, R. I. E. 1988. Ecological changes in Chesapeake Bay: Are theythe result of overharvesting the American oyster, Crassostrea virgi-nica? Pp. 536–546 in Understanding the Estuary: Advances in Ches-apeake Bay Research. Proceedings of a Conference, March 29–31,1988, Baltimore, MD, M. P. Lynch and E. C. Krome, eds. ChesapeakeResearch Consortium Publication 129, Chesapeake Research Consor-tium, Solomons, MD.


Newell, R. I. E, V. S. Kennedy, and K. S. Shaw. 2007. Comparativevulnerability to predators, and induced defense responses, of easternoysters Crassostrea virginica and non-native Crassostrea ariakensisoysters in Chesapeake Bay. Mar. Biol. 152: 449–460.

NRC (National Research Council). 2004. Nonnative Oysters in theChesapeake Bay, National Academies Press, Washington, DC.

Porter, E. T., J. C. Cornwell, and L. P. Sanford. 2004. Effect ofoysters Crassostrea virginica and bottom shear velocity on benthic-pelagic coupling and estuarine water quality. Mar. Ecol. Prog. Ser.271: 61–75.

Qasim, T., C. Ford, M. B. Bush, X. Hu, and B. R. Lawn. 2006. Effectof off-axis concentrated loading on failure of curved brittle layerstructures. J. Biomed. Mater. Res. Appl. Biomater. 76B: 334–339.

Qasim, T., C. Ford, M. B. Bush, X. Hu, K. A. Malament, and B. R.Lawn. 2007. Margin failures in brittle dome structures: relevance tofailure of dental crowns. J. Biomed. Mater. Res. Appl. Biomater. 80B:78–85.

R Core Team. 2013. R: A Language and Environment for StatisticalComputing. R Foundation for Statistical Computing, Vienna. [Online].Available: http://www.r-project.org/ [2013, September 6].

Rothschild, B. J., J. S. Ault, P. Goulletquer, and M. Heral. 1994.

Decline of the Chesapeake Bay oyster population: a century of habitatdestruction and overfishing. Mar. Ecol. Prog. Ser. 111: 29–39.

Sayegh, A. F., and S. B. Dong. 1970. Analysis of laminated curvedbeams. J. Eng. Mech. Div. 96: 471–482.

Singh, G., J. D. Block, and S. Rebach. 2000. Measurement of thecrushing force of the crab claw. Crustaceana. 73: 633–637.

Steimle, F. W., and C. Zetlin. 2000. Reef habitats in the Middle AtlanticBight: abundance, distribution, associated biological communities, andfishery resource use. Mar. Fish. Rev. 62: 24–42.

Ullmann, C. V., U. Wiechert, and C. Korte. 2010. Oxygen isotopefluctuations in a modern North Sea oyster (Crassostrea gigas) com-pared with annual variations in seawater temperature: implications forpalaeoclimate studies. Chem. Geol. 277: 160–166.

White, M. E., and E. A. Wilson. 1996. Predators, pests, and competi-tors. Pp. 559–580 in The Eastern Oyster, Crassostrea virginica, V. S.Kennedy, R. I. E. Newell, and A. F. Eble, eds. Maryland Sea GrantPublication, College Park, MD.

Yoon, Y., A. S. Mount, K. M. Hansen, and D. C. Hansen. 2009.Electrolyte conductivity through the shell of the eastern oyster using afour-electrode measurement. Bioelectrochemistry 156: 169–176.

Appendix

Differences in density and observed variations in hardness within Crassostrea ariakensis

Appendix Figure 1. Two microindentations from the same Crassostrea ariakensis shell. We believe theimage on the left is from the chalky layer, and the image on the right is from the non-chalky foliated layer. Theindentation shown in Panel A is almost 4 times larger than the indentation shown in Panel B. These largeindentations, possibly indicating chalky deposits, were not observed in Crassostrea virginica.


shell hardness and compressive strength of the eastern ... · c. virginica than for c. ariakensis...

Documents