Canadian Journal of Zoology. The right choice: predation pressure drives shell selection decisions in hermit crabs

The right choice: predation pressure drives shell selection decisions in hermit crabs Journal: Canadian Journal of Zoology Manuscript ID cjz-2017-0023.r2 Manuscript Type: Article Date Submitted by the Author: 10-Oct-2017 Complete List of Authors: Arce, Elsah; Universidad Autonoma del Estado de Morelos, ; Universidad Nacional Autonoma de Mexico, Cordoba, Alejandro; Universidad Nacional Autonoma de Mexico, Departamento de Ecología Evolutiva, Instituto de Ecología Keyword: Calcinus californiensis, hermit crab, gastropod shell, mistake, Shelter choice

Page 1 of 18 Canadian Journal of Zoology 1 The right choice: predation pressure drives shell selection decisions in hermit crabs 2 3 E. Arce 1,2* and A. Córdoba-Aguilar 2* 4 5 6 7 1 Laboratorio de Acuicultura, Departamento de Hidrobiología, Centro de Investigaciones Biológicas, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos, México. Tel: +52-77-73162354. E- mail: elsah.arce@uaem.mx 8 9 10 11 2 Departamento de Ecología Evolutiva, Instituto de Ecología, Universidad Nacional Autónoma de México, Apdo. P. 70-275, Ciudad Universitaria, 04510, México, D F., México. Tel: +52-55-56229003. E-mail: acordoba@iecologia.unam.mx 12 13 * Corresponding authors: 14 elsah.arce@uaem.mx & acordoba@iecologia.unam.mx 15 1

Page 2 of 18 16 The right choice: predation pressure drives shell selection decisions in hermit crabs 17 18 E. Arce and A. Córdoba-Aguilar 19 20 Abstract 21 22 23 24 25 Several prey species use refuges to avoid predation. Prey need to abandon and shift between refuges. However, during such shifting, prey can be vulnerable to predators. We hypothesize that predator presence may induce prey to make mistakes in choosing their refuge. We tested this by using hermit crabs Calcinus californiensis Bouvier, 1898 that were induced to shift to a new empty gastropod shell (three different species: Columbella sp. Lamarck, 1799, Nerita scabricosta Lamarck, 1822, and Stramonita biserialis 26 27 (Blainville, 1832)) in the absence and presence of a natural crab predator, Eriphia squamata Stimpson, 1860, an efficient shell-crushing predator. We expected that when a predator was present, hermit crabs would a) 28 29 30 31 32 33 inspect fewer shells and/or b) change to a shell that is either too heavy to allow escape or unfit in size to accommodate the hermit crab. While the first prediction was met, the second prediction was supported only when S. biserialis shells were used. Thus, in the presence of a predator, hermit crabs prioritize escaping via selecting lighter shells, which would allow the crab to move faster. We conclude that predator presence may induce prey to make mistakes in refuge selection, suggesting that this has severe consequences in future predatory events. 34 35 Keywords: shelter choice, mistake, gastropod shell, Calcinus californiensis, hermit crab 36 37 2

Page 3 of 18 Canadian Journal of Zoology 38 Introduction 39 40 41 42 43 44 45 46 47 48 Predators are a major selection pressure on prey, thus there has been selection for general adaptations to avoid detection, capture and ingestion in many prey species. Some examples of prey adaptations are alterations in activity, avoidance of sites occupied by predators, and an increase in their use of refuges (Brönmark and Miner 1992; Sih et al. 1992; Skelly 1994). It is well known that a refuge increases prey survival rate (i.e. Bertness 1981; Shervette et al. 2004). However, prey animals must find new refuges occasionally to avoid fitness costs in terms of growth (Lankford et al. 2001; Heithaus and Dill 2002) and reproduction (Spencer 2002; Auld and Houser 2015). This shifting between refuges exposes the animal to the cost of becoming exposed to predators at least temporarily. When facing such a situation, prey may make mistakes in choosing a new refuge, as prey must direct their attention to not only inspecting potential refuges but also to avoid capture by the predator. This means, for example, a reduced time to inspect potential refuges with the 49 remaining time devoted to detecting and avoiding the predator. 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 Hermit crabs are good study subjects to test the effect of predator threat on decision-making during refuge shifting (Hazlett 1971). These animals are characterized by a soft abdomen that requires protection from environmental stress and predation (Resse 1969; Vance 1972). Shell use is dependent on shell availability (Turra and Leite 2002; Arce and Alcaraz 2011) and preference (Meireles et al. 2008; Arce and Alcaraz 2012). The size and type of selected shells have direct effects on crab fitness. The proper size and type of shell may allow greater clutch sizes (Childress 1972; Elwood et al. 1995), higher growth rates (Asakura 1992; Alcaraz et al. 2015), lower risk of predation (Angel 2000; Arce and Alcaraz 2013) and less environmental stress (Bulinski 2007; Argüelles et al. 2009). It is also related to its morphological structure and dimensions; for example, tight shells offer more protection from being pulled out by predators (Arce and Alcaraz 2013), while thick shells are effective against shell-crusher predators (Bertness and Cunningham 1981). Nevertheless, hermit crabs frequently need to move into a new shell. During shell shifting, hermit crabs have to search, inspect, handle and change shells. These different steps can take time during the process of discarding and trying different shell sizes. It is during this move that hermit crabs run the risk of being predated on. Thus, the influence of a predator threat may induce hermit crabs to make mistakes in shell selection given the reduced time because of avoiding predation. To our knowledge, the effect of predation 3

Page 4 of 18 65 66 67 68 69 70 71 72 73 74 75 threat on the chance of making mistakes in shell selection has not been examined in hermit crabs or other animals. Although it is well known that predation risk can ultimately affect hermit crab preference and shell change (Mima et al. 2003; Briffa et al. 2008; Arce and Alcaraz 2013), these consequences can occur after a crab makes the mistakes. Thus, we have herein examined shell selection in the face of predation risk in hermit crabs and during shell shifting. For this, we used Calcinus californiensis Bouvier, 1898 hermit crabs and Eriphia squamata Stimpson, 1860 crab predators. The latter is a natural predator of C. californiensis that crushes the shell to reach the hermit crab (Bertness and Cunningham 1981). According to this predation strategy, hermit crabs should select light shells so that they can escape more effectively (Alcaraz and Arce 2017). Assuming that such crushing predators induce prey to make more mistakes, we therefore predicted that during the presence of a predator, crabs would a) inspect and/or change fewer shells, and b) select a shell that is not appropriate in terms of weight and/or size. 76 77 Materials and methods 78 79 Collection and maintenance 80 81 82 83 84 85 86 87 88 89 90 91 The animals used did not involve endangered or protected species. All applicable international, national and/or institutional guidelines for the care and use of animals were followed. Hermit crabs (C. californiensis) and predators (E. squamata) used for experimental trials were collected by hand in an intertidal zone at Troncones, Guerrero, Mexico (17 47 16 N; 101 44 17 W) in November and December 2014. Crabs were transported to the laboratory and maintained in individual containers to avoid shell exchange. Animals were removed from their shells and measured (shield length in mm and hermit crab mass in mg). We only used males of both hermit crabs and predators in our trials. For both, only animals that had all their appendages and shells with no clear damage or epibionts were used. All animals were kept under a natural light:dark photoperiod with running seawater (35 PSU) at an approximate temperature of 28 C. Crabs were fed with sinking pellets ( New Spectrum) once a day. Predators were maintained in individual containers and fed with hermit crabs 24 h prior to experiments. In all tests described below, different hermit crabs were used. On completion of trials, all animals were returned to the sea. 4

Page 5 of 18 Canadian Journal of Zoology 92 93 Shell selection during predator absence and presence 94 95 96 97 98 99 100 101 102 A hermit crab was placed with its original shell, when this was one of the most commonly occupied shells by C. californiensis (Arce and Alcaraz 2011, 2012). For instance, the one it was occupying when captured in the field, either Columbella sp. Lamarck, 1799, Nerita scabricosta Lamarck, 1822 or Stramonita biserialis (Blainville 1832) and with 20 other empty shells of the same species the crab was using in a 20-L individual glass tank for 24 h. The 20 shells offered varied in size. Each shell was previously weighed (Shell Mass, ShM), measured (using four morphometric variables: shell length, ShL; shell width, ShW; shell aperture length, ShAL and shell aperture width, ShAW) and marked with different color combinations. Direct observations and measurements were recorded every 6 hours. In this time, we recorded two behavioral aspects of the hermit crabs under these conditions (1 the number of shell inspections). A shell inspection took place 103 104 when the hermit crab held a shell with appendices for review, but discarded it (2 the number of shell changes). A shell change took place when the hermit crab abandoned its own shell and moved into a new 105 shell. We assumed that a hermit crab preferred a shell when it occupied it for as long as 24 h. 106 107 108 109 110 111 We then carried out a similar experiment in which E. squamata predators were introduced (Zipser and Vermeij 1978). These predators were fed with hermit crabs 24 h before the experiment started and were then placed in a 40-L glass tank. This tank was divided into two halves (20 L each) using a rigid and transparent mesh (5 mm). Water was allowed to circulate through the tank so that the chemical cues from E. squamata reached the hermit crab. We again recorded the number of shell inspections, shell changes and shell selection for 24 h. 112 113 Shell preference and the right choice 114 115 116 117 Shell preference in crabs occupying Columbella sp, N. scabricosta, and S. biserialis was assessed using the following combinations: hermit crabs in Columbella sp. in the absence of a predator (n = 15) and in the presence of a predator (n = 17); hermit crabs in N. scabricosta in the absence of a predator (n = 16) and in the presence of a predator (n = 14); and hermit crabs in S. biserialis in the absence of a predator (n = 15) and in 5

Page 6 of 18 118 119 120 121 122 123 the presence of a predator (n = 16). The shell selected by the crab after it remained in it for a 24-hour period was considered preferred. The number of mistakes was recorded during the shell selection process. A mistake was made when a hermit crab was occupying a shell out of its preferred weight and/or size (more than 10% tight or more than 10% loose; see Arce and Alcaraz 2012). The result of the preference for shells according to predator absence/presence was compared to determine whether the presence of a predator affects the decision. 124 125 Statistical analyses 126 127 The numbers of shell inspections and shell changes according to predator absence/presence were compared using a chi-square test (χ 2 ). Relationships between crabs and gastropod shells sizes were established by 128 129 regression analyses, where x = hermit crab size (M) and y = shell size (ShM or ShAW; Arce and Alcaraz 2012). The preferred shell was determined using this regression, which was performed for each shell species. 130 131 132 133 The slopes and elevations were compared using ANCOVA analysis of covariance to estimate the differences between the shell size selected under predator absence and under predator presence. We also used ANOVA to compare the number of mistakes in predator absence versus predator presence. All reported values are means ± SE. Statistical analyses were conducted using Statistics 7. 134 135 Results 136 137 138 139 140 141 142 Hermit crabs inspected fewer gastropod shells when a predator was present than when it was absent (ᵡ2(2,1) = 28.46, p < 0.001; Fig. 1). This situation prevailed independently of the gastropod species (F (5,56) = 15.91, p < 0.001; Fig. 1). Hermit crabs showed more shell changes during predator absence than during predator presence (ᵡ2(2,1) = 8.91, p < 0.01; Fig. 2) and when hermit crabs occupied Columbella sp. shells during predator absence, compared to occupation of S. biserialis and N. scabricosta (F (5,56) = 18.50, p < 0.001; Fig. 2). Finally, when the predator was present, shell changes did not differ among the three shell species (Fig. 2). 6

Page 7 of 18 Canadian Journal of Zoology 143 144 145 146 147 148 149 150 151 Hermit crabs selected different shell sizes or weights when a predator was absent than when the predator was present (Fig. 3). Specifically, hermit crabs occupying Columbella sp. (F (2,30) = 15.62 p < 0.001, Fig. 3a), N. scabricosta (F (2,28) = 18.86 p < 0.001, Fig. 3b) and S. biserialis (F (2,29) = 16.78 p < 0.001, Fig. 3c) preferred smaller or lighter gastropod shells when a predator was present. The number of mistakes during the shell selection process was different according to the predator absence/presence condition and shell species occupancy. When hermit crabs occupied Columbella sp. and N. scabricosta shells in the absence of a predator they made more mistakes than when the predator was present (Table 1). However, when hermit crabs occupied S. biserialis shells and the predator was absent, they made fewer mistakes than when the predator was present (Table 1). 152 153 Discussion 154 155 We postulated that although the presence of a predator may elicit behavioral responses to avoid predation in prey, this pressure might also affect prey decisions. In concordance with our first prediction, hermit crabs 156 157 158 159 160 161 162 163 164 165 166 167 168 169 made fewer changes when a natural predator was present. Moreover, our results supported our second prediction, since crabs made more mistakes when a natural predator was present in terms of choosing a suboptimal shell but only when using S. biserialis shells. Previous studies also found that animals modify their behavior in the presence of predators; for example, hermit crabs modified their shell selection behavior in the presence of a predator (Rotjan et al. 2004; Arce and Alcaraz 2013; Alcaraz and Arce 2017) and other crustaceans showed less activity when predation risk was high (Hazlett 1996; Percor and Hazlett 2003). These results, in addition to the results of the present study, imply that hermit crabs can derive less information about resource value, in this case, the utility of the shell as an effective refuge, in the face of predator pressure. Related to predator presence, one antipredator response that hermit crabs show is a retraction inside the occupied shell (Briffa and Elwood 2001), which is a generalized response in animals using refuges (Sih et al. 2004). However, one immediate consequence of retracting is that hidden animals lose feeding and reproductive opportunities (Reaney 2007). For example, when the predator Carcinus maenas (Linnaeus 1758) is present, the prey Nucella lapillus (Linnaeus 1758) reduces foraging activity, as it remains hidden for longer (Vadas et al. 1994). The same result can occur even in prey that does not rely on refuge (Matassa and Trussell 7

Page 8 of 18 170 171 2014). This is the case, for example, of lepidopteran larvae that reduce their foraging rate when the perceived level of predation risk is high (Berger and Gotthard 2008). 172 173 174 175 176 177 178 179 180 Prey-catching strategies by predators play key roles in how well prey can determine an effective refuge size (Walters and Wethey 1996). The risk of predation in hermit crabs is strongly related to shell size relative to crab size. According to previous studies (i.e. Arce and Alcaraz 2013), we used this relationship as a tool to determine whether crabs had made the right choice of refuge size, because the fit is essential for hermit crabs during predation attacks. In relation to this, C. californiensis deals with two types of natural predators in their habitat: peelers and crushers. Peelers (e.g. the crab Arenaeus mexicanus (Gerstaecker 1856)) pull out the crab by directly cutting and removing parts of the crab, while crushers (i.e. the predator we used, E. squamata) break the shell with their chelae (Bertness and Cunningham 1981). Interestingly, each predator style performs better with certain species of shell than others. For example, some shells allow the hermit crab 181 182 to retreat completely and become unreachable to a peeler predator (Angel 2000). However, such shells are ineffective if the predator is a crusher. With a crusher, the hermit crab has to flee before its shell is grabbed. In 183 184 185 186 187 188 189 190 191 this latter circumstance, a hermit crab does better if it bears a light shell, as it allows faster movements. According to this context, we found that C. californiensis crabs selected tighter or lighter shells when they perceived a threat by E. squamata. This is concordant with results from previous studies that suggest light shells allow hermit crabs to move faster (Herreid and Full 1986; Alcaraz and Arce 2017) and escape effectively when faced with a predator. Thus, the use of light shells by C. californiensis may be adaptive to avoid attacks by the crusher E. squamata (Alcaraz and Arce 2017). However, when choosing light shells, the crab made more mistakes if it faced a predator when occupying S. biserialis (the preferred shell; Arce and Alcaraz 2012). This shell species may have some additional yet undetected costs, which may only arise when a predator is present. Future studies can be conducted to clarify this. 192 193 194 195 196 Individuals may benefit from modifying their resource searching behavior when they perceive predation risk. However, this perception may induce them to make mistakes during shell selection. An incorrect choice and/or suboptimal shelter does not provide the individual with the benefits of counteracting a predatory attack. In this context, our study demonstrated that resource selection in the face of predation might induce mistakes by restricting inspection to fewer shells, which may have severe consequences for future 8

Page 9 of 18 Canadian Journal of Zoology 197 198 predatory events. To our knowledge, we have provided the first evidence of the occurrence of such mistakes and the costs prey incur from these mistakes when a predator is present. 199 200 Acknowledgements 201 202 203 We thank the DGAPA-UNAM Postdoctoral Program for providing a full scholarship to the first author. We thank Karla Kruesi for technical support in the field and the laboratory, Suntai Sotola and Manuel De La Torre for the drawings for Figures, and Guillermina Alcaraz for logistic and academic support. 204 205 References 206 207 Alcaraz, Z.G., Chávez-Solís, C.E., and Kruesi, C.K. 2015. Mismatch between body growth and shell preference in hermit crabs is explained by protection from predators. Hydrobiologia, 743(1): 151 156. 208 doi:10.1007/s10750-014-2029-8. 209 210 Alcaraz, Z.G., and Arce, U.E. 2017. Predator discrimination in the hermit crab Calcinus californiensis: tight for shell breakers, loose for shell peelers. Oikos, 126(9): 1299 1307. doi:10.1111/oik.03742. 211 212 Angel, J.E. 2000. Effects of shell fit on the biology of the hermit crab Pagurus longicarpus (Say). J. Exp. Mar. Biol. Ecol. 243: 169 184. doi:10.1016/s0022-0981(99)00119-7. 213 214 Arce, U.E., and Alcaraz, Z.G. 2011. Shell use by the intertidal hermit crab Calcinus californiensis at different levels of the intertidal zone. Sci. Mar. 75: 121 128. doi:10.3989/scimar.2011.75n1121. 215 216 217 Arce, U.E., and Alcaraz, Z.G. 2012. Shell preference in a hermit crab: comparison between a matrix of paired comparisons and a multiple-alternative experiment. Mar. Biol. 159: 853 862. doi:10.1007/s00227-011- 1861-x. 218 219 Arce, U.E., and Alcaraz, Z.G. 2013. Plasticity of shell preference and its antipredatory advantages in the hermit crab Calcinus californiensis. Can. J. Zool. 91(5): 321 327. doi.org/10.1139/cjz-2012-0310. 9

Page 10 of 18 220 221 222 Argüelles, T.A., Álvarez, F., and Alcaraz, Z.G. 2009. Shell architecture and its relation to shell occupation by the hermit crab Clibanarius antillensis under varying conditions of wave action. Sci. Mar. 73(4): 717 723. doi:0.3989/scimar.2009.73n4717. 223 224 Asakura, A. 1992. Population ecology of the sand-dwelling hermit crab Diogenes nitidimanus Terao. V. Ecological implications in the pattern of molting. J. Crustac. Biol. 12: 537 545. doi:10.2307/1548835. 225 226 Auld, J.R., and Houser, R. 2015. Age dependent effects of predation risk on reproductive success in a freshwater snail. Evolution, 69(10): 2793 2798. doi.org/10.1111/evo.12769. 227 228 Berger, D., and Gotthard, K. 2008. Time stress, predation risk and diurnal nocturnal foraging trade-offs in larval prey. Behav. Ecol. Sociobiol. 62: 1655 1663. doi: 10.1007/s00265-008-0594-4. 229 230 Bertness, M.D. 1981. Interference, exploitation, and sexual components of competition in a tropical hermit crab assemblage. J. Exp. Mar. Biol. Ecol. 12: 537 545. doi:10.1016/0022-0981(81)90070-8. 231 232 Bertness, M.D., and Cunningham, C. 1981. Crab shell-crushing predation and gastropod architectural defense. J. Exp. Mar. Biol. Ecol. 50: 213 230. doi:10.1016/0022-0981(81)90051-4. 233 234 Briffa, M., and Elwood, R.W. 2001. Motivational change during shell fights in the hermit crab Pagurus bernhardus. Anim. Behav. 62: 505 510. doi:10.1006/anbe.2001.1764. 235 236 237 Briffa, M., Haskell, P., and Wilding, C. 2008. Behavioural colour change in the hermit crab Pagurus bernhardus: reduced crypticity when the threat of predation is high. Behaviour, 145: 915 929. Available from http://www.jstor.org/stable/40295870. 238 239 Brönmark, C., and Miner, J.G. 1992. Predator-induced phenotypical change in body morphology in crucian carp. Science, 258(5086): 1348 1350. doi: 10.1126/science.258.5086.1348. 240 241 Bulinski, K.V. 2007. Shell-selection behavior of the hermit crab Pagurus granosimanus in relation to isolation, competition, and predation. J. Shellfish Res. 26(1): 233 239. doi:10.2983/0730-8000. 10

Page 11 of 18 Canadian Journal of Zoology 242 243 Childress, J.R. 1972. Behavioral ecology and fitness theory in a tropical hermit crab. Ecology, 53: 960 964. doi:10.2307/1934316. 244 245 246 Elwood, R.W., Marks, N., and Dick, J.T.A. 1995. Consequences of shell-species preferences for female reproductive success in the hermit crab Pagurus bernhardus. Mar. Biol. 123(3): 431 434. doi: 10.1007/BF00349221. 247 248 Hazlett, B.A. 1971. Influence of rearing conditions on initial shell entering behavior of a hermit crab (Decapoda, Paguridea). Crustaceana, 20:167 170. Available from http://www.jstor.org/stable/20101773. 249 250 Hazlett, B.A. 1996. Assessments during shell exchanges by the hermit crab Clibanarius vittatus: the complete negotiator. Anim. Behav. 51: 567 573. doi: 10.1006/anbe.1996.0060. 251 252 Heithaus, M.R., and Dill, L.M. 2002. Food availability and tiger shark predation risk influence bottlenose dolphin habitat use. Ecology, 83(2): 480 491. doi: 10.1890/0012-253 9658(2002)083[0480:FAATSP]2.0.CO;2. 254 255 Herreid, C.F., and Full, R.J. 1986. Locomotion of hermit crabs (Coenobita compressus) on beach and treadmill. J. Exp. Mar. Biol. Ecol. 120: 283 296. 256 257 258 Lankford, T.E., Billerbeck, J.M., and Conover, D.O. 2001. Evolution of intrinsic growth and energy acquisition rates. II. Trade offs with vulnerability to predation in Menidia menidia. Evolution, 55(9): 1873 1881. doi:10.1111/j.0014-3820.2001.tb00836.x. 259 260 Matassa, C.M., and Trussell, G.C. 2014. Prey state shapes the effects of temporal variation in predation risk. Proc. R. Soc. 281: 20141952. doi:10.1098/rspb.2014.1952. 261 262 263 Meireles, A.L., Biagi, R., and Mantelatto, F.L. 2008. Influence of prior experience on shell selection by the white spot wrist hermit crab Pagurus criniticornis (Crustacea: Paguridae). Hydrobiologia, 605: 259 263. doi:10.1007/s10750-008-9305-4. 264 265 Mima, A., Wada, S., and Goshima, S. 2003. Antipredator defence of the hermit crab Pagurus fiholi induced by predatory crabs. Oikos, 102(1): 104 110. doi: 10.1034/j.1600-0706.2003.12361.x. 11

Page 12 of 18 266 267 268 Pecor, K.W., and Hazlett, B.A. 2003. Frequency of encounter with risk and the tradeoff between pursuit and antipredator behaviors in crayfish: a test of the risk allocation hypothesis. Ethology, 109: 97 106. doi: 10.1046/j.1439-0310.2003.00834.x. 269 270 Reaney, L.T. 2007. Foraging and mating opportunities influence refuge use in the fiddler crab, Uca mjoebergi. Anim. Behav. 73: 711 716. doi:10.1016/j.anbehav.2006.05.022. 271 272 Reese, E.S. 1969. Behavioral adaptations of intertidal hermit crabs. Am. Zool. 9(2): 343 355. doi:http://dx.doi.org/10.1093/icb/9.2.343. 273 274 275 Rotjan, R.D., Blum, J., and Lewis, S.M. 2004. Shell choice in Pagurus longicarpus hermit crabs: does predation threat influence shell selection behavior? Behav. Ecol. Sociobiol. 56: 171 176. doi: 10.1007/s00265-004-0770-0. 276 Shervette, V.R., Perry, H.M., Rakocinski, C.F., and Biesiot, P.M. 2004. Factors influencing refuge occupation 277 by stone crab Menippe adina juveniles in Mississippi sound. J. Crustac. Biol. 24(4): 652 665. 278 279 Sih, A., Bell, A., and Johnson, J.C. 2004. Behavioural syndromes: an ecological and evolutionary overview. Trends Ecol. Evol. 19: 372 377. doi: 10.1016/j.tree.2004.04.009. 280 281 Sih, A., Kats, L.B., and Moore, R.D. 1992. Effects of predatory sunfish on the density, drift and refuge use of stream salamander larvae. Ecology, 73: 1418 1430. doi: 10.2307/1940687. 282 283 Skelly, D.K. 1994. Activity level and the susceptibility of anuran larvae to predation. Anim. Behav. 47: 465 468. doi:10.1006/anbe.1994.1063 47:465-468. 284 285 Spencer, R.J. 2002. Experimentally testing nest site selection: fitness trade-offs and predation risk in turtles. Ecology, 83(8): 2136 2144. doi:10.1890/0012-9658(2002)083[2136:etnssf]2.0.co;2. 286 287 Turra, A., and Leite, F.P.P. 2002. Shell utilization patterns of a tropical intertidal hermit crab assemblage. J. Mar. Biol. 82(1): 97 107. doi.org/10.1017/s0025315402005210. 12

Page 13 of 18 Canadian Journal of Zoology 288 289 290 Vadas, R., Burrows, M., and Hughes, R.N. 1994. Foraging strategies of dogwhelks, Nucella lapillus (L.): interacting effects of age, diet and chemical cues to the threat of predation. Oecologia, 100: 439 450. doi:10.1007/bf00317866. 291 292 Vance, R.R. 1972. The role of shell adequacy in behavior interactions in hermit crabs. Ecology, 53: 1075 1083. doi:10.2307/1935419. 293 294 295 Walters, L.J., and Wethey, D.S. 1996. Settlement and early post-settlement survival of sessile marine invertebrates on topographically complex surfaces: the importance of refuge dimensions and adult morphology. Mar. Ecol. Prog. Ser. 137:161 171. Available from http://www.int-res.com/journals/. 296 297 Zipser, E., and Vermeij, G.J. 1978. Crushing behavior of tropical and temperate crabs. J. Exp. Biol. Ecol. 31: 155 172. doi:10.1016/0022-0981(78)90127-2. 298 13

Page 14 of 18 299 Figure captions: 300 301 302 303 Figure 1. Number of shell inspections by hermit crabs occupying different gastropod shell species during predator absence/presence scenarios. Different letters indicate significant differences between treatments, while different numbers indicate differences between gastropod shell species (p < 0.001). 304 305 306 307 308 Figure 2. Number of shell changes by hermit crabs occupying different gastropod shell species according to predator absence/presence scenarios. Different letters indicate significant differences between treatments, while different numbers indicate differences between gastropod shell species (p < 0.001). 309 310 311 Figure 3. Regression slopes according to shell size preference by hermit crabs: A) Columbella sp., B) Nerita scabricosta and C) Stramonita biserialis. R 2, correlation coefficient (p < 0.001). 312 313 314 315 316 317 318 319 320 321 322 14

Page 15 of 18 Canadian Journal of Zoology 323 324 Table 1. Analysis of variance testing the number of mistakes hermit crabs made when getting into a new shell species under predator absence/presence scenarios. 325 Gastropod shell Predator Mistakes SE df F P stimulus (n) Columbella sp. Absence 54 0.94 Presence 25 0.89 1 10.33 <0.01 N. scabricosta Absence 52 0.64 Presence 31 0.89 1 7.35 <0.01 S. biserialis Absence 37 0.47 Presence 66 0.20 1 64.69 <0.001 326 327 15

Page 16 of 18 328 Figure 1 329 330 331 16

Page 17 of 18 Canadian Journal of Zoology 332 Figure 2 333 334 335 17

Page 18 of 18 336 Figure 3 Shell Aperture Length (mm) 5 4 3 2 1 Predator absence Predator presence y = 4.4101x + 2.0394 R² = 0.731 y = 2.8644x + 1.4922 R² = 0.3676 A 0 0 0.05 0.1 0.15 0.2 Hermit crab mass (g) 1.5 Predator absence y = 1.9333x + 0.337 R² = 0.7767 Predator presence y = 1.6898x + 0.0946 B R² = 0.721 1 Shell mass (g) 0.5 0 0 0.1 0.2 0.3 0.4 Hermit crab mass (g) 5 Predator absence y = 4.1657x + 0.9666 R² = 0.7928 C 4 Predator presence y = 3.6477x + 0.4541 R² = 0.8671 Shell mass (g) 3 2 1 0 0 0.2 0.4 0.6 0.8 1 Hermit crab 18 mass (g)