“main” — 2010/11/4 — 12:09 — page 833 — #1 Anais da Academia Brasileira de Ciências (2010) 82(4): 833-842 (Annals of the Brazilian Academy of Sciences) ISSN 0001-3765 www.scielo.br/aabc Characterization of the occupied shells by the hermit crab Clibanarius vittatus (Decapoda, Diogenidae) at Baixio Mirim tideflat, Guaratuba Bay, southern Brazil SARA R. SAMPAIO and SETUKO MASUNARI Departamento de Zoologia, Universidade Federal do Paraná, Centro Politécnico Caixa Postal 19020, 81541-990 Curitiba, PR, Brasil Manuscript received on May 19, 2009; accepted for publication on August 31, 2009 ABSTRACT A characterization of the occupied shells by the hermit crab Clibanarius vittatus was carried out. Hermit crabs were collected in the intertidal zone, during the low spring tide monthly from April 2005 to March 2006. They were sexed and their cephalothoracic shield length (CL) was measured. Shells were identified, dried, weighed and the aperture length (AL) and width (AW) were measured. 1187 crabs were collected (949 males, 216 females and 22 intersexes), which occupied 12 species of gastropod shells. Stramonita haemastoma, Olivancillaria urceus and Dorsanum moni- liferum made up 96.55% of the total shell species. Male hermit crabs attained significantly larger sizes than females; therefore, males occupied a wider spectrum of shells in size and weight. A stronger correlation ratio was obtained between CL and AW of S. haemastoma. Last whorl with a rounded shape and a spacious inner area is a common feature of all shell species most frequently occupied by this hermit crab where it occurs. The successful establishment of C. vittatus at Baixio Mirim is mainly due to the appropriately shaped and wide range of size of S. haemastoma shells that were most often occupied by the hermit crabs of the studied population. Key words: Guaratuba Bay, shell availability, shell occupation, shell size and shape. INTRODUCTION and Leite 2001, Meireles et al. 2003). When this avail- ability is restricted in the habitat, hermit crabs are com- Hermit crabs are anomuran crustaceans that use empty pelled to occupy shells that are not always appropriate in shells of gastropods as a shelter. These shells can be size, form or integrity, and this restriction affects repro- acquired from the habitat if empty and available; if not, duction (Childress 1972, Bertness 1981, Hazlett 1989, by removing dead or dying mollusks, once hermit crabs Bertini and Fransozo 2000, Hazlett et al. 2005, Turra are unable to dislodge alive gastropods from their shells 2005), growth (Markham 1968, Bertness 1981, Osorno (Rittschof 1980, Bertness 1981, Tricarico and Gherardi et al. 1998, Bertini and Fransozo 2000) and body mor- 2006). The gastropod shells perform several important phology (Turra and Leite 2002). roles in the life of these crabs, mainly the protection of Secondly, the occupation of a specific shell is in- their uncalcified abdomen against eventual mechanical fluenced either by the interactions among hermit crabs, shocks and dryness. such as intra-specific competition for the shell, orby The use of the specific shell species by a hermit other resources (Bertness 1980, 1981, Gherardi and Nar- crab is firstly related to the availability of these shells done 1997, Dominciano and Mantelatto 2004, Sant’Anna in the environment (see revision in Vance 1972, Turra et al. 2006b), coexistence of different hermit crabs spe- Correspondence to: Sara Regina Sampaio cies (Kellog 1977, Bertness 1980, Gherardi and Nardone E-mail: [email protected] An Acad Bras Cienc (2010) 82 (4) “main” — 2010/11/4 — 12:09 — page 834 — #2 834 SARA R. SAMPAIO and SETUKO MASUNARI 1997, Floeter et al. 2000, Turra 2003, Sant’Anna et al. Hermit crabs occur individually or in groups of 2006b), predation (Bertness 1982, Rotjan et al. 2004), 5-10 crabs either in the vegetated areas or in non-veg- presence of gastropods in the community (Bertness 1980, etated ones in Baixio Mirim tideflat, from where they Turra and Leite 2001, Sant’Anna et al. 2006a), previous were collected during low spring tides, monthly from experience (Hazlett 1993, Hahn 1998, Meireles et al. April/2005 to March/2006. Each monthly sample con- 2008, Alcaraz and Kruesi 2009) and formation of ag- sisted of nearly a hundred crabs collected with the fol- gregations (Hazlett 1981, Turra and Leite 2000a). The lowing shell size frequency distribution: 25% of indi- hermit crab ability in detecting odours of died or dy- viduals occupying small shells, 50% medium and 25% ing gastropods was also reported (Kratt and Rittschof big ones. This procedure aimed to avoid collections bi- 1991, Chiussi et al. 2001). ased towards more visible shells because hermit crabs Likewise, morphometric characteristics of shells have a gregarious habit, and a sampling based on the influence their occupation by hermit crabs. Among area or CPUE would be improper. Furthermore, such these characteristics are weight (Hahn 1998, Domincia- number was sufficient or more than sufficient to repre- no and Mantelatto 2004, Turra and Leite 2004), size sent populations of hermit crabs in other studies in the (Bertini and Fransozo 2000, Turra and Leite 2004), open- Brazilian coast (Fransozo and Mantelatto 1998, Floeter ing size (Botelho and Costa 2000, Bertini and Fransozo et al. 2000, Turra and Leite 2000b, 2001, Sant’Anna 2000, Sant’Anna et al. 2006a) and architecture and vol- et al. 2006a). ume (Gherardi and Nardone 1997, Floeter et al. 2000, Hermit crabs were packed in plastic bags and kept in a thermal box; at the laboratory, they were preserved Shih and Mok 2000). in freezer until their handling. They were extracted by Clibanarius vittatus (Bosc, 1802) has a wide geo- hand from the occupied shells but, when necessary, it graphical distribution in western Atlantic, from the east- was done with the help of a small vice. After this, the ern of the United States to West Indies, including Gulf crabs were preserved in alcohol 70% while shells were of Mexico, and from Venezuela to southern Brazil, in dried up indoors for 24h. Hermit crabs were identified, Santa Catarina State (Melo 1999). In spite of the wide sexed, and the length (CL) of the cephalothoracic shield distribution, the ecological aspects of this species were was obtained with a 0.01mm precision digital caliper. only reported in the USA in North and South Carolina, Sexes were identified according to the position of gen- Florida and Gulf of Mexico, and in Brazil these aspects ital openings; intersex individuals had both male and are restricted to the São Paulo State coast (Turra 2003, female genital openings. Empty shells were identified 2005, Turra and Leite 2001, 2002, 2004, Sant’Anna et and weighted (SWE) with an electronic scale of 0.01g al. 2006a, b). The present study is the first research precision, and their aperture length (AL) and width about the relationships of the hermit crab C. vittatus and (AW) were measured with the same caliper. Only the its occupied shell from the Guaratuba Bay, Paraná State, most occupied shells were used for comparative anal- southern Brazil. ysis and, therefore, those having less than 5% occupa- tion were excluded. MATERIALS AND METHODS Krukal-Wallis was used for statistical comparison Baixio Mirim is a 6.300m2 tideflat located within Gua- of the dimensions among the different shell species, and ratuba Bay and isolated from the continent by a nar- for comparison of these dimensions in the occupied row channel of 15-20m width. Most of its substrate is shells by different sexes (these dimensions had not a nor- covered by the smooth cordgrass Spartina alterniflora mal distribution), the Dunn’s test was used to locate the Loisel, but it is also provided with sandy and muddy source of variations. The χ 2 was used to compare the surface especially in the bordering areas. In spite of ur- shell occupation frequencies between males and females. ban influence, this tideflat is inhabited by hundreds of Correlation Indices of Spearman were calculated for all fiddler crabs that constitute the main preys for seabirds correlations among dimensions: only complete and un- that land on the sandbanks during low tides (Masunari et broken shells were used in this analysis. All tests had a al. 2005). confidence of 95%. An Acad Bras Cienc (2010) 82 (4) “main” — 2010/11/4 — 12:09 — page 835 — #3 CHARACTERIZATION OF THE OCCUPIED SHELLS Clibanarius vittatus IN GUARATUBA BAY 835 RESULTS non-ovigerous and ovigerous females. AL and AW of the shells occupied by the latter ones had no significant A total of 1187 hermit crabs Clibanarius vittatus were difference from those occupied by intersexes (Kruskal- collected, out of which 949 were males, 204 non-ovi- Wallis, p < 0.05) (Table III). gerous females, 12 ovigerous females and 22 inter- The three most frequently occupied shell species sexes. They occupied 12 species of gastropod shells, were common among these sexes. Furthermore, a pat- with the highest frequency of occupation in Stramonita tern in the shell use was observed: S. haemastoma shells haemastoma (Linnaeus, 1767) with 64.61%, followed were occupied by almost all hermit crab sizes, O. urceus by Olivancillaria urceus (Röding, 1798) with 26.47%, by medium-sized crabs and D. moniliferum by smaller and Dorsanum moniliferum (Valenciennes, 1834) with ones. However, each sex occupied these shells with dis- 5.48%. The remainder shells were occupied with less tinct frequencies: 29.29% males used O. urceus, while than 1% each one, and altogether they did not sum up 15.58% females and 18.18% intersexes did it. In con- 4% – Achatina fulica (Bowdich, 1822), Buccinanops trast, only 4.52% males, 8.33% females and 13.64% lamarckii (Kiener, 1834), Cymatium parthenopeum (Von intersexes occupied D. moniliferum (Fig. 3). Salis, 1793), Chicoreus (Siratus) senegalensis (Gmelin, In the correlation analyses between hermit crabs 1791), Olivancillaria steeriae (Reeve, 1850), Olivancil- CL and the shell species dimensions, S.