THE VELIGER

The Veliger 48(4):235-242(January 25,2007) O CMS, Inc.,2005

Shell Microstructure of the Patellid Gastropod Collisella scabra (Gould): Ecological and Phylogenetic Implications

SARAH E. GILMAN* Section of Evolution and Ecology, and Center for Population Biology, University of California, Davis, Davis, cA 95616-8755

Abstract. Shell microstructure has a long history of use in both taxonomic and ecological research on molluscs. I report here on a study of the microstructure of Collisella scabra, also known as Macclintockia scabra and Lottia scabra. I used a combination of SEM and light microscopy of acetate peels, whole shells, and shell fragments to examine the shell layers and microstructure. Regular growth bands were not present in most shells examined. Several shells showed multiple bands of myostracum, which indicate periods of extreme rates of size change, and may be evidence of abiotic stress. This study also suggests that shell layers previously described as "modified foliated" are "irregular complex crossed lamellar," with both fibrous and foliated second order structures. The presence of fibrous shell structures, in addition to other shared morphological characters noted by previous authors, suggests an affinity with the Lottiidae rather than the Nacellidae.

INTRODUCTION scabra shells, and 2) to verify the earlier shell microstructure descriptions, including phylogenetic Shell microstructure has a long history of use in both implications. taxonomic and ecological research on molluscs. Shell growth bands are commonly used as records of individual growth (Frank, 1975; Hughes, 1986; Arnold MATERIALS AND METHODS et al., 1998) and for reconstructing environmental conditions (Rhoads &LuIz,l980; Jones, 1981;Kirby et Collection and Preparation of Shells al., 1998). Shell microstructure has also been used for taxonomic identification (Kennish et al., 1998) and as I collected 10-20 snails from each of the three field evidence of shared ancestry in systematic work (Kool, sites listed in Table 1. All three sites are rocky intertidal (sensu 1993; Schneider & Carter, 2001). In revising the benches on semi-protected outer coast Ricketts systematics of the patellogastropod genus Collisella, et &1., 1985). At each site, I collected snails from primarily Lindberg (1986) used shell microstructure characters wave sheltered areas, such as surge channels protected described by MacClintock (1967) to place Collisella or the wave sides of outcrops or ridges. I specifically scabra in a separate family from the rest of the species searched for snails with a minimum of shell erosion in the genus. Although Lindberg never formally and collected a range of sizes at each site. In the lab, I dissected each from its published a new taxonomy for this species (see snail shell. Lindberg, 1986), Kozloff (1987, 1996) contains Lind- berg's intended systematics and the name Macclin- Whole Shells tockia scabra has appeared in the published literature (e.g., Smith et a1., 1993). More recently, Fuchigami & To examine the muscle scar and adjacent layers on Sasaki (2005) re-examined the shell microstructure of the inner surface of the shell, I soaked several shells in 44 patellogastropod species, and updated several of 57a sodium hypochlorite for one hour and then rinsed MacClintock's (1967) original descriptions, including them under running water. I then examined the inner C. scabra.In this study I reexamined the microstructure surface of these shells under a dissecting microscope. of C. scabra with two goals: 1) to explore the use of To determine the mineral structure of the shell layers shell growth patterns as a way to age individual C. exposed on the inner surface, I treated two shells with Feigl Stain (Feigl, 1937; Freidman,1959), which stains * PresentAddress: Friday Harbor Laboratories, IJniversity of aragonitic areas black, but leaves calcitic areas un- Washington, Friday Harbor, WA 98250, e-mail: gilmans@ changed. I also used three of these bleached shells in u.washington.edu scanning electron microscopy (see below). 4 Page 236 The Veliger, Vol. 48, No.

Iable I

Location of samples collected for analysis.

of snails Protocol Collection Site Location Collection Date Number l0 Acetate Fort Bragg, California (MFB) 39.282'N,123.803"W 05107l0l 0610610r l2 Feigl Stain. SEM l2 Acetate Shelter Cove. California (SCV) 40.022'N. 124.074"W 05/06/01 l5 Acetate, SEM Devil's Gate. California (DVG) 40.407"N.124.390"W 06109101

follow the definitions rn Acetate Peels describing shell structures Carter et al. (1990). Shells for acetate peels were embedded in clear epoxy The myostracum is a narrow band of irregular - resin (EPON 828 resin and DTA hardener, Miller simple prismatic structure (Figure 2)' Adjacent and the Stephenson, Danbury CT), and sectioned through exterior to the myostracum is the m + I layer apex along the midsagittal plane using an Isomet low (Figure 2), a very narrow band of branching crossed I ground and speed saw (Buehler Ltd., Lake Bluff, IL). lamellar structure. This layer is rarely visible in acetate Carter & etched the sections and made peels following peels, and much more visible in the SEMs or by direct mm thick acetate for the Ambrose (1989). I used 2 examination of the interior surface of bleached shells problems with curling and tearing peels, which reduces under a dissecting microscoPe. with much thinner acetate (Carter commonly reported The majority of the shell consists of the m + 2 and 1989). After 12-24 hr' I examined peels & Ambrose, m 1 layers, which contain a similar first order a compound microscope at 25x to 250x. under structure (Figures 2-5), best described as irregular Photographs of specimens were made with a 35 mm complex crossed lamellar (Carter et al., 1990). SEM body mounted onto the microscope. camera revealed the second order structure to be highly variable, consisting of long fibers (i.e., "fibrous Electron Microscopy prismatic," Figure 5) of varying width that grade into foliated sheets (i.e., "regularly foliated," Figure 4). I examined whole shells, shell fragments, and epoxy MacClintock (1967) indicated a possible m + 3 layer, embedded sections with scanning electron microscopy not describe it. In general, the shells examined (SEM). Whole shells and shell fragments were pre- but did viously bleached as described above and air dried for several days before use. Shell fragments were generated by gently tapping a whole shell with a hammer. Three APICAL from those used for embedded specimens were selected EXTERIOR the acetate peels. These were polished with aluminum oxide and re-etched in 57o HCI for 15 sec before SEM' All samples were prepared for SEM following Carter & Ambrose (1989). Once thoroughly air-dried, they were INTERIOR mounted onto specimen support stubs using silver m m-1 MARGINAL suspension paste (Ted Pella Inc., Redding' CA) and mf1 sputter-coated with gold (Pelco Model SC-7, Pelco International, Redding CA). Samples were viewed in ANTERIOR a Philips XL 30 Scanning Electron Microscope (FEI Co., Hilsboro OR) oPerated at 10 KV'

RESULTS

I identified four layers in the shell of C. scabra (Figure I ). Feigl staining revealed the muscle scar (myostracum) to be aragonite and the remaining shell POSTERIOR layers, calcite. Following MacClintock ( 1961), I iden- layers by their position relative to the tify these Figure l. Diagram of shell layers in C. sc'qbra. (A) Cross- are myostracum. Layers interior to the myostracum section of shell along sagittal axis, after MacClintock (1967' labeled with negative numbers, and shell layers exterior Figure l).(B) Shell layers as viewed on the ventral surface of to the myostracum are labeled with positive numbers the shell. after MacClintock (1967, Figure 82)' The shaded by "m" is the myostracum (muscle scar)' (Figure I ). Unless otherwise indicted all terms used in area indicated S. E. Gilnran.2005 Page 237

yi'

FigLrre2. SEM of'sa-tlittalsection ctl-siutrplcDVG 2A. Thc sllell rvas cntbcc'lcleclin epoxy apcl sectic-rleclbcli re

crossecllantellar. Scale bar is l0 micron for this study were too hi-ehlyeroclecl to cletermineif an DISCUSSION m+3layerexisted. Becauseof the irre_qularityof the The -soalsol- this str"rclywere two-fblcl: I ) to explore the nt + " und nt I layers, it is difflcLrlt to see consistent growth lines. luseof sl-rellgrowth patterns its erwuy to age individual Growtl-r discontinuities (i.e.. "growtl'r bands") are Colli.sallu.s't'uhru sl-rells. arnd 2) to verify earlier occasionally observed in acetate peels ( Fi-cLrre6). clescriptions and tlteir tuxonon-ric in-rplicatior-rs.Shell althougl-rthey were ilpparellt only in portions of shell growth berncisarc colllt-ron in other r-nollusksand are cross-sectionsand ltever consistently across the entire ofien usecl in flshcries research to er-qeindividuals shell.Other shellshad no banclsat al (e.g.. tri_uure7). (Rlroacls& Lutz. 1980).Banclin_s is occasictnallyevident The banding appears to be causeclby chan_eesin the in C. .st'tthru,but generally only in larger individr"rals orientation of tl-rehigher orcler structures (e.-q..m - | runclnot throughout the entire shell.Most shellsshowed layer of Figlrres 2 ancl 9). no growth banclsat all. The bands appear to be fbrmed Irregr-rlaritieswere also present in the myostrircLn-n. by chan-eesin tl-reorientatior-r of the tnd order-eler-nents In most shells. the myostralcLln-lcor-rsistecl of a sir-rgle of the m l ar-rclnt + t luyers. It is unclear how such band that expancleclwith the growth of thc- shell; orientartior-rclttrr-rgcs ntigltt relatc to annunl or seasonal however. in some serrnplesthele zlppears to be lnore patterns of growth. Thus. visr-ralshell bands are than one band of ntyostrucLlntancl/or the myostracultl Lrrrlikelyto t'rer,rsefirl fbl aging indiviclr"ralC. .stubru. appearsto dor-rbleback on itself (Figures 8 9). More intrigLring is thc r-rccusior-tulobservation of Page238 The Veliger, Vol. 48, No. 4

multiple bands of myostracum within a single shell (Figure 5). MacClintock (1967, plate 7) reported a similar pattern in the shells of Lottta gigantea, which he refers to as evidence of "an earlier extended period." The multiple bands likely occur when changes in the size of the animal (shrinkage or expansion) are rapid relative to the rate of growth of the rest of the shell; and suggest that individuals may be repositioning them- selves within the shell over their lifetime, possibly as the result of starvation. Multiple bands were usually only observed in the largest shells examined. These individuals occupy the highest vertical distribution of the species in the intertidal (Sutherland, 1970; Haven, 1973). Such areas show extreme seasonal patterns in food supply, and long periods of starvation are likely (Sutherland, l9l0; Cubit, 1984). Sutherland (1970) observed a seasonal loss of body mass in these snails, which might change the position of the body within the shell and relocate the myostracum to a more apical position. This is very different from the breaks reported in other mollusks, where shell growth stops during reproduction (Rhoads &Lut2,1980). In general, the changes in the position of the myostracum did not appear frequently enough to suggest a seasonal or annual pattern ofsize change; and thus are unlikely to be useful as measures of indivi- dual age. However, they might be useful as evidence of extreme environmental stress over longer time intervals. Perhaps more importantly, the description of shell structure for Collisella scabra differs materially from both the original description by MacClintock (1967) and the more recent description of Fuchigami & Sasaki (2005), and the phylogenetic position of this species should be reconsidered. At the time of MacClintock's work, Collisella scabra was placed within the Acmaei- dae, but was removed by Lindberg (1986) as part of a larger revision of the northeastern Pacific Acmaeidae. Although Lindberg never formally published a new taxonomy for this species, he intended to place it in the Family Nacellidae, genus Maccltntockia (Kozloff, 1987, 1996; Lindberg pers. comm.). Because Lindberg's (1986) reclassification of the northeastern Pacihc species of Acmaeidae was heavily influenced by MacClintock's (1967) work, the redescription of C. scabra's shell microstructure warrants a reexamination of its phylogenetic position. My observations of Collisella scabra's shell micro- - MacClintock's (1967) earlier Figures 3-5. Secondary structure of the m + 2 and m I structure dilfer from layers revealed by SEM of fractured shells. (3) Fracture along description in both the number of shell layers described a transverse axis showing cross sections of fibers. The and their composition. I identified five of the six shell structures vary from rod like at the upper part of the image layers mentioned in MacClintock (1967); but, like foliated at the lower portion. Scale bar is to lath or Fuchigami and Sasaki (2005), I was unable to identify l0 microns. (4-5) Details of secondary structure from other an m + 3 layer. The shells I examined may have been locations of the same shell. Scale bar is 5 microns. (4) Foliated + layer. Within the sheets, (5) blades or laths. too highly eroded to retain the m 3 S. E. Giln-ran.2005 Page 239

i:.{:Si , .l.:.., |&''

,-t .rp.*.

fi, t,? ; .,{t, ..w &; :S,.: i?:f Lt'r

...la._::.l ; .,'i

FigLrres6 7. Acctatc pccls ol- thc apcx ttl- twt-rclill-ercrttsttt.ttplcs vicu,ecl unclcr a conrpouucl n.ricrc.rscopc.((r) Sanr;llc DVG -5A. banding is clcarly visitrlc.(7) SanrplcSCV llA. no banclsr,'isiblc.

- Pase 240 The Veliger, Vol. 48, No. 4

(8) Two separate lityers of myostracllm are Figures 8-9. SEM of an etched sagittal section, posterior slope of sample CPM 5,{. micron. (9) Enlargemeut of A. sczrlebar is indicated by arrows. Tl-reblack box indicates the area represented in B, scale bzrr is 100 20 micron. Two intersecting bancls of myostracuttt elre clearly visible along the dizr-eonal. S. E. Gilman, 2005 Page 241 five layers I observed, I also identified different primary "completely unrelated to all other speciesof the family and secondary structures. In particular, MacClintock Acmaeidae" (MacClintock, 1967p.82) becauseof the (1967 p. 76) consideredthe m + 2 and m - I layers to "modified foliated" structuresfound in its m+ 2 and m be "modified foliated or possibly modified fibrillar." I layers. This study demonstrates that, although Fuchigami & Sasaki (2005) identified the m - I layer as calcitic, the second order structure of these layers varies "irregular complex crossedfoliated" and the m + 2layer from fibrous prismatic to foliated. Becausethey appear as "irregular fibrous foliated." I have identified both distinct from any other microstructures observed in the these layers as "irregularly complex crossed lamellar." Lottiidae or Patellidae,they are autapomorphies of C. Also, both MacClintock (1967 p.76) and Fuchigami & scabra, and provide no phylogenetic information. Sasaki (2005) labeled the m + I layer as "concentric C. scabra shares many other morphological char- crossed lamellar," while I describe it as "branching acters, including gill and radular morphology, with the crossedlamellar." Lottiidae that are not common to other patellogas- These differences are due to both terminological and tropod families (MacClintock, 1967; Lindberg, l98l). methodological differences between the two studies. Furthermore, molecular analyses(Simison, 2000; Simi- MacClintock (1967) used primarily thin sections and son & Lindberg, 2003) also support a hypothesis of examination of whole shells to diagnose shell layers; shared ancestry between C. scabra and members of the whereas, awtate peels and SEMs provide much better genus Lottia. Thus it is likely that C. scabra belongs, detail of structure. Additionally, descriptive terminol- with its former congeners, in the genus Lottia, family ogy of skeletal microstructure remains highly variable Lottiidae. among researchers(Carter et al., 1990), and many of the terms used by MacClintock (1967) are not Acknowledgments. This research was supported by NSF grant commonly in use today. In the caseof the m + I layer, OCE-9906741 to R. Grosberg and by a DOE Graduate Research Environmental Fellowship to S. Gilman. all three studies identified a crossed lamellar structure. I am indebted to J. Carter for sharing his wealth of knowledge on (1967) MacClintock and Fuchigami & Sasaki (2005) shell microstructure, acetate peels, and microscopy. SEM describedit as "concentric," referring to the orientation work was conducted in the Department of Medical Pathology of the frrst order structures relative to the shell margin. at UC Davis, with technical support from K. Butler-DeRose. This orientation is consistent with the definition of J. Carter, R. Grosberg, M. Frey, G. Herbert, and G. Vermeij "branching crossedlamellar" (Carter et al., 1990). provided many helpful comments on earlier drafts of this manuscript. MacClintock (1967) termed the primary structure of both the m + 2and m - I layers "modified foliated or possibly modifred fibrillar," which is not a commonly REFERENCES CITED used term in shell microstructure studies. Based on ARNoLD,W.S., T. M. BERT,I. R. QurruyeR & D. S. JoNes. MacClintock's (1967) description of these layers, 1998. Contemporaneous deposition of annual growth Carter et al. (1990 p.6a\ considered that they were bands in Mercenaria mercenaria (Linnaeus), Mercenaria 'oflibrous campechiensls(Gmelin), and their natural hybrid forms. likely to be either prismatic" or "irregular Journal of Experimental Marine Biology and Ecology complex crossedlamellar." Fuchigami & Sasaki (2005) 223:93-109. describedseparate microstructures for the two layers. I CARTER,J. G. & W. W. ANasnosr. 1989. Techniques for have identiflred microstructures similar to each of studying molluscan shell microstructure. Pp. 101-119 in Fuchigami & Sasaki's (2005) structures. Their "irreg- R. M. Feldman, R. E. Chapman & J. T. Hannibal (eds.), ular fibrous foliated" is similar to Figure 4 of this Paleotechniques.Dept. of Geological Sciences,IJniv. of Tenn., Knoxville: Knoxville, study, and their "irregular complex crossed foliated" is TN. CARTER,J.G., K. BeNoeI-, V. DE BuRrner'nl, S. J. CARLsoN, similar to Figure 5. However, I observed these two J. CasreNgr, M.A. CnnNsHRw,J. E. Der-rNcwRTER,H. structures intergrading within the same shell layer FneNctr-r-oN-Vrrrllot, J. GrnnuDrE, F. J. MBurqrnn,H. (Figure 3). I have described both layers as "irregular MurvEr, A. oe Rrcqlrs, J. Y. Srnr, A. B. SurrH, J. complex crossed lamellar" based on the deflrnition of WENor, A. Wrllnus & L. ZvrseRsrnc. 1990.Glossary (Carter et al., 1990),which encompassessecond order of skeletal biomineralization. Pp. 6O9472 in J. G. Carter (ed.), structures ranging from fibers to planar lamellae (i.e., Skeletal Biomineratzation: Patterns, Processesand Evolutionary Trends. Van Nostrand Reinhold: New sheets).Carter et al. (1990)consider "irregular complex York, NY. crossed foliated" a variant of "irregular complex CUBIT,J. D. 1984.Herbivory and the seasonalabundance of crossedlamellar" specific to foliated secondary struc- algae on a high intertidal rocky shore. Ecology 65:1904- tures. Thus I have described both the m - I and m + 2 1917. layers as "irregular complex crossed lamellar" with FEIGL, F. 1937. Qualitative analysis by spot tests, inorganic a variable secondary structure. and organic applications. Nordemann Publishing Com- pany Inc.: New York, NY. ix, 400 pp. MacClintock (1967) considered foliated shell struc- FRANT, P. W. 1975. Latitudinal variation in the life history tures to be characteristic of the Patellidae and absent in features of the black turban snail Tegula funebralis the Acrnaeidae.Thus he concluded that C. scabra was (Prosobranchia:Trochidae). Marine Biology 3 I : 18l-192. Page 242 The Veliger, Vol. 48, No. 4

FRrIDIIA.N, G. M. 1959. Identification of carbonate minerals LrNoseRc, D. R. 1981. Acmaeidae: Gastropoda Mollusca. by staining methods. Journal of Sedimentary Petrology Boxwood Press: Pacific Grove, Calif. r,n,122 pp. 29:87-97. Lrr.roseRc, D. R. 1986. Name changes in the "Acmaeidae". FucHrcevrr, T. & T. Sasaru. 2005. The shell structure of the The Veliger 29:142-148. Recent Patellogastropoda (Mollusca: Gastropoda). Pale- MacCrrrqrocK, C. 1967. Shell structure of patelloid and ontological Research 9:143-168. bellerophontoid gastropods (Mollusca). Peabody Muse- fIAvEN, S. B. 1973. Competition for food between the um of Natural History. Bulletin 22:i-I40. intertidal gastropods Acmaea scabra and Acmaea digitalis. RHoAos, D. C. & R. A. Ltsrz. 1980. Skeletal growth of Ecology 54:143-151. aquatic organisms: biological records of environmental HucHEs, R. N. 1986. A functional biology of marine change. Plenum Press: New York. xiii, 750 pp. gastropods. Johns Hopkins University Press: Baltimore, Rrcrrrrs, E. F., J. CarvrN, J. W. HsocpsrH & D. W. MD. 245 pp. Prutups. 1985.Between Pacific tides. Stanford University JoNEs, D. S. 1981. Annual growth increments in shells of Press: Stanford, Calif. xxvi, 652 pp. Spisula solidissima record marine temperature variability. Scrrxsroen, J. A. & J. G. CanreR. 2001. Evolution and 2Ll:165-167. Science phylogenetic significance of cardioidean shell microstruc- M. A. TaN R. A. Lvrz. 1998. Shell KENNTSH, J., S. & ture (Mollusca, Bivalvia). Journal of Paleontology 75: microstructure of mytitds (Bivalvia) from deep-sea 607-&3. hydrothermal vent and cold-water sulfide/methane seep SnraIsoN,W. B. 2000. Evolution and phylogeography of New environments. Nautilus I 12:84-89. World gastropod faunas. Ph.D. Integrative Biology KrRBy, M. X., T. M. SoI*IIAI & H. J. Speno. 1998. Stable University of California: Berkeley. vli, 202 pp. isotope sclerochronology of Pleistoceneand recent oyster StvttsoN,W. B. & D. R. Lnrnssnc.2OO3. On the identity of shells (Crassostrea vir ginica). Palaios I 3:560-569. Lottia strigatella (Carpenter, 1864) Koot, S. P. 1993. Phylogenetic analysis of the Rapaninae @atellogastropoda: (Neogastropoda: Muricidae). Malacolo g1a3 5:I 55J59. Lottiidae). The Veliger 46:l-19. KozLoFF, E. N. 1987. Marine invertebrates of the Pacific SvrrrH, A.M., W. M. KrBn & S. JonNsnN. 1993.The effect of Northwest. University of Washington Press: Seattle. depth on the attachment force of limpets. Biological 5ll pp. Bulletin 184:338-341. KozLoFF, E. N. 1996. Marine invertebrates of the Pacific SutHeRtAND, J. P. 1970. Dynamics of high and low Northwest. University of Washington Press: Seattle. populations of the limpet Acmaea scabra (Gould). 539 pp. Ecological Monographs 40: I 69-1 88.

i ilI il