COEXISTENCE SHAPES ANTIPREDATOR DEFENSES

3Present address: European Commission, Joint Research Centre, Institute for Environment and

Sustainability, Water Resources Unit, Ispra, Italy. E-mail: aluisanunes@gmail.com

Rapid evolution of constitutive and inducible defenses against an invasive predator 1

ANA L. NUNES1,2,3

, GERMÁN ORIZAOLA2, ANSSI LAURILA

2, AND RUI REBELO

1 2

3

1Centro de Biologia Ambiental, D.B.A., Faculdade de Ciências da Universidade de Lisboa, 4

Lisboa, Portugal 5

2Animal Ecology/Department of Ecology and Genetics, Evolutionary Biology Centre, Uppsala 6

University, Uppsala, Sweden7

NUNES ET AL.

2

Abstract 8

Invasive alien predators can impose strong selection on native prey populations and induce rapid 9

evolutionary change in the invaded communities. However, studies on evolutionary responses to 10

invasive predators are often complicated by the lack of replicate populations differing in 11

coexistence time with the predator, which would allow determining how prey traits change during 12

the invasion. The red swamp crayfish Procambarus clarkii has invaded many freshwater areas 13

worldwide, with negative impacts for native fauna. Here, we examined how coexistence time 14

shapes antipredator responses of the Iberian waterfrog (Pelophylax perezi) to the invasive crayfish 15

by raising tadpoles from five populations differing in historical exposure to P. clarkii (30 years, 20 16

years or no coexistence). Tadpoles from non-invaded populations responded to the presence of P. 17

clarkii with behavioral plasticity (reduced activity), whereas long-term invaded populations 18

showed canalized antipredator behavior (constant low activity level). Tadpoles from one of the 19

long-term invaded populations responded to the crayfish with inducible morphological defenses 20

(deeper tails), reflecting the use of both constitutive and inducible antipredator defenses against the 21

exotic predator by this population. Our results suggest that, while naive P. perezi populations 22

responded behaviorally to P. clarkii, the strong predation pressure imposed by the crayfish has 23

induced the evolution of qualitatively different antipredator defenses in populations with longer 24

coexistence time. These responses suggest that strong selection by invasive predators may drive 25

rapid evolutionary change in invaded communities. Examining responses of prey species to 26

biological invasions using multiple populations will help us better forecast the impact of invasive 27

predators in natural communities. 28

Keywords: coexistence time, invasive predator, plastic and constitutive defenses, population, 29

rapid evolution, predator recognition 30

31

NUNES ET AL.

3

Introduction 32

Predation is an important selective force acting on the fitness of prey organisms, and selection on 33

prey populations for improved ability to escape or deter predation is often strong (Lima and Dill 34

1990). Prey exhibit a wide variety of antipredator defenses involving modifications in behavior, 35

morphology and life-history that reduce their vulnerability to predators (Lima and Dill 1990, 36

Tollrian and Harvell 1999, Benard 2004). These defenses can be expressed constitutively, or be 37

inducible and only expressed when predators are present (Tollrian and Harvell 1999). Constitutive, 38

or canalized, defenses are common in habitats under constant and high predator pressure (Clark 39

and Harvell 1992, Moran 1992, Edgell et al. 2009). On the contrary, inducible, or plastic, defenses 40

evolve in habitats with spatially or temporally variable predation risk and are favored when 41

reliable cues from predation are available (Padilla and Adolph 1996). Inducible defenses have a 42

fitness advantage when predators are present, but incur costs in the absence of predators, 43

preventing them from becoming constitutive (Clark and Harvell 1992, Tollrian and Harvell 1999). 44

While constitutive and plastic defenses are often considered as independent alternatives, selection 45

for both types of defenses can act simultaneously on different traits, and the expression and 46

integration of the different defenses may vary among populations (DeWitt et al. 1999, Poitrineau 47

et al. 2004, Bourdeau 2012). 48

Successful predator recognition and effective avoidance strategies typically occur when predator 49

and prey share a common evolutionary history (Strauss et al. 2006, Sih et al. 2010). Invasive alien 50

predators are spreading over new environments at an unprecedented rate, resulting in new 51

predator-prey interactions between organisms sharing no evolutionary history (Thompson 1998, 52

Strauss et al. 2006). Invaded prey populations often suffer dramatic declines and even extinctions 53

due to their inability to respond with appropriate antipredator defenses (Cox and Lima 2006, 54

NUNES ET AL.

4

Strauss et al. 2006, Salo et al. 2007, Sih et al. 2010). However, by imposing strong selection 55

pressure upon native organisms, invasive predators may also induce rapid evolutionary changes, 56

through which prey may evolve the ability to recognize and respond to them (Schlaepfer et al. 57

2005, Strauss et al. 2006, Sih et al. 2010). For rapid evolution to occur, selection pressure imposed 58

by the exotic predator has to be strong and consistent, and native populations have to be 59

genetically variable in susceptibility to the predator (Strauss et al. 2006). Indeed, rapid adaptive 60

evolution in native prey species as a response to introduced predators has been reported in several 61

taxa including mollusks, crustaceans, fish, amphibians and reptiles (Kiesecker and Blaustein 1997, 62

O’Steen et al. 2002, Losos et al. 2004, Freeman and Byers 2006, Fisk et al. 2007). 63

However, some prey are able to innately detect and elicit adaptive responses to stimuli of novel 64

predators in the absence of a common evolutionary history (Epp and Gabor 2008, Peluc et al. 65

2008, Sündermann et al. 2008, Rehage et al. 2009). This ability may be due to the latter’s 66

phylogenetic or phenotypic similarity to native predators co-occurring with the prey (Epp and 67

Gabor 2008, Rehage et al. 2009). Prey may also exhibit generalized antipredator responses, 68

whereby exposure to predation risk induces the expression of defenses, regardless of previous 69

direct experience with a particular predator (e.g. response to a general fish odor or to a large 70

moving object; Magurran 1990, Gherardi et al. 2011). This can be facilitated if detection of 71

predators relies on general predation cues, such as cues associated with the predator’s diet or alarm 72

cues released by startled or attacked prey (Sündermann et al. 2008, Ferrari et al. 2010b). 73

Alternatively to innate responses, prey may learn to identify and respond to alien predators 74

through experience by pairing novel predator scents with cues released by attacked or consumed 75

conspecifics and hence learn to ‘label’ exotic predators as a threat (reviewed in Ferrari et al. 76

2010b). 77

NUNES ET AL.

5

The potential for the evolution of antipredator responses may vary among populations due to 78

factors such as the intensity of selection, the amount of genetic variation within prey populations 79

or the genetic architecture of the traits (Strauss et al. 2006). Studies using single or few invaded 80

populations may, thus, fail to fully show the complexity of the evolutionary responses of natives to 81

introduced predators. Therefore, designing studies using populations with differing co-82

evolutionary history with the invasive predator may be crucial for better forecasting the long-term 83

impacts of biological invasions (see e.g. Kiesecker and Blaustein 1997, Trussell and Smith 2000, 84

Edgell et al. 2009). 85

Freshwater environments, which have been widely invaded by exotic species, are particularly 86

vulnerable to introduced predators (Cox and Lima 2006). One of the most conspicuous non-native 87

predator species introduced worldwide is the red swamp crayfish (Procambarus clarkii), native to 88

southern USA and northern Mexico (Gherardi 2006). This invasive species was first recorded in 89

Europe in the Iberian Peninsula in 1973 and has rapidly spread throughout the area (Holdich et al. 90

2009). P. clarkii is an extraordinarily successful invader, often altering the dynamics of the 91

ecosystems it colonizes (e.g. Geiger et al. 2005). Crayfish are efficient predators of larvae of many 92

amphibian species (e.g. Cruz and Rebelo 2005, Cruz et al. 2008). 93

Tadpoles of the Iberian waterfrog (Pelophylax perezi) exhibit typical plastic behavioral (reduced 94

activity level) and morphological (shorter and deeper headbodies and deeper tails) antipredator 95

responses when exposed to chemical stimuli of native aquatic predators (Richter-Boix et al. 2007, 96

Polo-Cavia et al. 2010, Gómez-Mestre and Díaz-Paniagua 2011, Nunes 2012, Nunes et al. 2013). 97

Although previous studies have not detected behavioral or morphological responses of P. perezi to 98

P. clarkii (Gómez-Mestre and Díaz-Paniagua 2011, Nunes 2012, Nunes et al. 2013), Gómez-99

Mestre and Díaz-Paniagua (2011) found that induced morphological defenses elicited by P. perezi 100

NUNES ET AL.

6

in the presence of a native larval dragonfly predator (a deeper and more pigmented tail) increased 101

tadpole survival in the presence of P. clarkii. 102

In this study, we investigated how coexistence time with P. clarkii influences the ability of P. 103

perezi to respond to this invasive predator. We compared behavioral and morphological responses 104

of five P. perezi populations characterized by different exposure times to P. clarkii: long (ca. 30 105

yr), medium (ca. 20 yr) or no coexistence time with the predator. Assuming that P. clarkii imposes 106

strong selection pressure upon P. perezi populations (Cruz and Rebelo 2005, Cruz et al. 2008), we 107

expect that tadpoles from long-term invaded populations will show antipredator defenses towards 108

the crayfish (such as alterations in morphology and/or a reduction in activity rates), while 109

individuals from non-invaded populations will show no responses to this predator. The medium-110

term invaded population should present intermediate responses, or at least respond differently 111

from the non-invaded populations. 112

113

Methods 114

Study species and populations 115

The Iberian waterfrog (P. perezi, see PLATE 1) is the most abundant and widely distributed 116

anuran in Portugal (Loureiro et al. 2008). Although inhabiting a wide variety of aquatic habitats, it 117

reproduces preferably in permanent water bodies (Almeida et al. 2001). We selected five 118

populations of P. perezi from southern Portugal, a region where P. clarkii is currently widespread 119

and where there are no present or historical records of native crayfishes (Almaça 1991). Two 120

populations were collected in the area of the Caia River, the region where P. clarkii was first 121

recorded in Portugal in 1979 (Ramos and Pereira 1981; Table 1, Fig. 1), and they constitute the 122

populations with longer coexistence time with P. clarkii in this study (10-15 P. perezi generations, 123

NUNES ET AL.

7

hereafter LC populations). A population with shorter coexistence time (hereafter SC) was 124

collected at the Melides River, an area that was invaded by the crayfish ca. 1990 (R. Rebelo pers. 125

obs.; Table 1, Fig. 1). Finally, we collected two P. perezi populations in the Monchique region, 126

which have never been in contact with P. clarkii (No Coexistence, hereafter NC; Table 1, Fig. 1). 127

The Monchique region is characterized by numerous natural springs, streams and rivers, where 128

topographic barriers and the distance to the crayfish introduction source may have prevented the 129

invasion by P. clarkii until now (Kerby et al. 2005, Cruz and Rebelo 2007). Repeated surveys 130

using baited traps in 2009 proved this area devoid of crayfish (A.L. Nunes and R. Rebelo, 131

unpubl.). Trapping and visual observations confirmed that dense P. clarkii populations coexisted 132

with P. perezi in LC and SC populations. All study populations were collected from slow-flowing 133

sections of streams harboring a rich fauna of native predators (insects, fish and snakes) or from 134

rice fields with immediate connection to the river. Although the Caia river harbors a more diverse 135

fish community (9 species) than the other two areas (5 species; Ribeiro et al. 2007), the former 136

community is composed of mostly detritivorous and insectivorous species not known to actively 137

prey on tadpoles. Overall, apart from the differences in crayfish invasion history, localities were 138

selected to be as identical as possible on other ecological parameters that could possibly influence 139

the traits studied here (e.g. vegetation structure, water temperature, shading, etc.). 140

Experimental procedure 141

During June and July 2009, we collected 8-12 freshly laid egg clutches of P. perezi at each of the 142

study populations (Table 1). The eggs were taken to the laboratory in the field station of Centro de 143

Biologia Ambiental in Grândola (38º06’N, 8º34’W), where the experiment took place. Each clutch 144

was divided in two and kept in 1L containers filled with well water until the start of the 145

experiment. Adult crayfish were caught in streams close to the field station using baited funnel 146

NUNES ET AL.

8

traps, kept in 28 × 20 × 14 cm containers and fed commercial fish food ad libitum until the start of 147

the experiment. 148

The experiment was a 5 x 2 factorial design with population origin crossed with two predator 149

treatments. When most of the larvae reached Gosner developmental stage 25 (complete absorption 150

of gills and active feeding; Gosner 1960), we divided them haphazardly into groups of 10 151

individuals, which were then allocated to the experimental containers (39 × 28 × 28 cm) filled with 152

10L of well water (day 1 of the experiment), where they remained throughout the whole 153

experiment. Containers were provided with an opaque suspended predator cage (diameter 85 mm, 154

length 16 cm) with top and bottom covered with fine mesh netting (1.5 mm), allowing the tadpoles 155

to receive chemical, but no visual or tactile cues from the crayfish. Half of the cages received one 156

crayfish (predator treatment), whereas the other half were left empty (no-predator treatment). Each 157

population/predator treatment combination was replicated once in each of five randomized blocks. 158

Crayfish were fed inside the predator cages with three P. perezi tadpoles every other day, and 159

tadpoles were fed commercial fish food and boiled lettuce ad libitum every three days. Water in 160

the containers was changed once a week. Water temperature was kept at 21.0 ºC (± 0.1) and the 161

photoperiod was 12L:12D. 162

Response variables 163

We recorded tadpole activity level at early (day 10 of the experiment) and mid (day 60 of the 164

experiment) larval development. Larval developmental stage of tadpoles at day 10 was considered 165

identical for all treatment combinations, and it did not differ among populations (P= 0.75) or 166

predation treatments (P= 0.66) at day 60 (Gosner range 33-36). The number of active tadpoles in 167

each tank, during a 10-s observation interval, was recorded five times between 9 and 12 a.m., the 168

NUNES ET AL.

9

observations being separated by at least 30 minutes. Tadpoles were considered active if they were 169

swimming, foraging or moving their tails. 170

We used geometric morphometrics tools to examine variation in tadpole morphology (Rohlf and 171

Marcus 1993, Zelditch et al. 2004). Morphology was recorded at day 60 of the experiment in five 172

haphazardly-selected individuals per container, by placing each tadpole in a small Plexiglas box 173

with a grid background and taking a side-view digital photograph. To minimize the effect of non-174

parallel tail positions, all the images were straightened with the straighten plugin in ImageJ 175

(http://rsb.info.nih.gov/ij, see Johansson et al. 2010 for details) and then loaded into MakeFan7 176

(Sheets 2009) to create a standardized template for digitizing the landmarks. For each tadpole, 19 177

landmarks representing an adequate coverage of the tadpole body shape were marked using the 178

image analysis software tpsDig2 (Rohlf 2008; Appendix 1; see e.g. Orizaola et al. 2012, 2013, for 179

a similar approach). In order to extract tadpole shape variables, landmark coordinates were 180

imported into tpsRelw (Rohlf 2007) and a Generalized least squares Procrustes Analysis was 181

performed in order to standardize the size, translate and rotate the configurations of landmark 182

coordinates (Rohlf and Slice 1990). TpsRelw allowed us to generate a mean (consensus) landmark 183

configuration for each experimental container, described by several components (relative warps), 184

which were later used as shape variables in the statistical analysis. The two first relative warps 185

(RW1 and RW2) explained most of the initial morphological variance (see Results) and were used 186

as dependent variables in the analyses. 187

Statistical analysis 188

We examined variation in antipredator responses for activity rates using a Generalized Linear 189

Mixed Model (GLMM) with binomial error distribution and a logit link function. We used time 190

period and predator treatment as fixed factors, time of coexistence with crayfish as a continuous 191

NUNES ET AL.

10

fixed factor and population as a random factor. We tested for all the possible interactions between 192

terms, but the 3-way interaction between time period, predator treatment and coexistence time was 193

removed from the final model because it was not significant (F1,89= 1.185, P= 0.279). Tank was 194

used as an additional random effect in order to account for potential overdispersion in the data. 195

The proportion of active tadpoles per container, estimated as the number of tadpoles active divided 196

by their total number in a container, was used as the dependent variable. This value was the 197

average of the five measurements recorded per container in each time period. LSD pairwise 198

comparisons were performed to inspect differences among populations. 199

We examined the effect of predator treatment on tadpole morphology across the different 200

populations using general linear model (GLM) analyses followed by Tukey HSD post-hoc tests. 201

Predator treatment was included as a fixed factor, time of coexistence as a fixed continuous 202

variable, and population origin as a random effect to account for the replication of related 203

genotypes. Since we found a nearly significant population × treatment interaction (see Results), 204

GLMs were performed for each population separately. Experimental block was initially used as a 205

random factor, but since this factor was never significant (P > 0.1) it was removed from the final 206

models. REML was used as estimation procedure. All analyses were performed using the software 207

package SPSS (version 21.0). 208

209

Results 210

A significant interaction between time period and predator treatment (F1,90= 13.3, P= 0.020) 211

indicated higher activity in the absence of crayfish as larval development proceeded (day 60, Fig. 212

2). We also found a significant interaction between time period and coexistence time with crayfish 213

(F1,90= 10.7, P< 0.001) reflecting that, in general, activity differs among populations with different 214

NUNES ET AL.

11

coexistence times as larval development advances, with activity increasing with lower coexistence 215

time with crayfish in day 60 of the experiment (Fig. 2). The evolutionary history of crayfish 216

contact affected the tadpoles' behavioral response to the presence of a fed crayfish (coexistence 217

time × predator treatment interaction: F1,90= 34.8, P< 0.001; Fig. 2). Overall, tadpoles from 218

populations with longer coexistence time with crayfish (LC1 and LC2) had very low activity 219

levels, and did not modify their activity in response to crayfish presence (pairwise comparisons, 220

P< 0.05; Fig. 2). When exposed to fed crayfish, tadpoles from populations without crayfish (NC1 221

and NC2) and populations coexisting for a short time with crayfish (SC) showed, overall, lower 222

activity rates than control tadpoles from the same populations (pairwise comparisons, P< 0.05; 223

Fig. 2). When looking at each population separately, only tadpoles from SC population had lower 224

activity when in the presence of fed crayfish early in larval development (day 10 of the 225

experiment, P= 0.003). As larval development advanced (day 60 of the experiment), tadpoles from 226

the two NC populations exposed to crayfish became less active than those from the respective 227

control treatments (P< 0.001), whereas activity of tadpoles from LC and SC populations was not 228

affected by treatment (Fig. 2). 229

Morphometric analyses on tadpole morphology yielded 34 relative warps, of which the first two 230

(RW1 and RW2) explained 54.78% of the total shape variation. RW1 accounted mainly for 231

constitutive differences between populations (P= 0.09), with populations coexisting with crayfish 232

(both LC and SC) having a more bulky body, with longer and deeper headbodies, expanded 233

particularly around the gut area (negative values of RW1; post-hoc tests, P< 0.05; Fig. 3a). For 234

RW1, no significant differences between predator treatments or a population × predator treatment 235

interaction were found (P> 0.523; Fig. 3a). For the second relative warp (RW2), there was a non-236

significant tendency towards a population × predator interaction (F4,36 = 2.1, P= 0.096). Additional 237

NUNES ET AL.

12

analyses showed that, in the presence of fed crayfish, tadpoles from population LC1 expanded 238

their tail region, developing deeper tail muscles and tail fins (F1,6 = 10.5, P= 0.018; Fig. 3b, 239

positive values of RW2), indicating morphological plasticity in response to crayfish in this 240

population. No evidence for alterations in morphology in response to crayfish presence was found 241

in any of the other populations (P≥ 0.1; Fig. 3b). 242

243

Discussion 244

We found that P. perezi populations differing in coexistence time with P. clarkii showed variation 245

in their antipredator responses to this exotic crayfish. Specifically, populations with a long 246

coexistence with P. clarkii (LC) exhibited a constitutive behavioral defense (maintenance of very 247

low activity levels even in the absence of predators) and LC1 population showed adaptive 248

morphological plasticity in the presence of fed crayfish. On the contrary, populations with shorter 249

(SC) or no (NC) coexistence time with crayfish showed plastic antipredator behavior when 250

exposed to the fed crayfish and lacked plastic morphological responses. These results highlight 251

that i) invasive predators can promote rapid evolution of adaptive antipredator responses in native 252

prey, ii) the responses in individual traits can be either constitutive or plastic, and iii) even naive 253

populations can respond adaptively to exotic predators. The considerable variation detected among 254

the P. perezi populations reflects the complexity of antipredator responses to P. clarkii, and 255

perhaps to invasive predators in general, and may partly explain why previous studies using either 256

a single population or two geographically close populations of P. perezi have failed to detect 257

antipredator responses in this predator-prey system (Gómez-Mestre and Díaz-Paniagua 2011, 258

Nunes 2012, Nunes et al. 2013). 259

NUNES ET AL.

13

Contrary to our predictions, populations of P. perezi with longer coexistence history with P. 260

clarkii did not alter their activity level when exposed to fed crayfish. Instead, tadpoles in these 261

populations maintained very low activity levels irrespective of predator treatment, suggesting the 262

evolution of a constitutive (canalized) behavioral defense. In general, constitutive defenses are 263

expected to evolve under strong uniform and prolonged predation risk (Moran 1992, Edgell et al. 264

2009). In our case, strong directional selection by P. clarkii towards low activity appears to have 265

promoted the loss of behavioral plasticity in LC populations, leading to permanently defended, 266

inactive tadpoles. Similar results were reported by Edgell et al. (2009), who showed that snails 267

from the native area of a crab predator show constitutive behavioral defenses against the predator, 268

whereas conspecific snails from areas recently invaded by this predator show behavioral plasticity. 269

A longer coexistence time or a more intense selection pressure may be required for this 270

constitutive response to evolve in SC populations. It is important to mention, though, that the 271

amphibian eggs were collected in the field shortly after laying and the early embryonic stages in 272

LC and SC populations may have been exposed to crayfish cues for a short time. While embryonic 273

learning of predators has been demonstrated in amphibians (Ferrari et al. 2010a), we find it 274

unlikely that this short early exposure would have affected antipredator responses during the larval 275

stage. Transgenerational effects, with mothers pre-conditioning larval strategies as a response to 276

predator presence (e.g. McGhee et al. 2012), may be another way of responding to invasive 277

predators, although this option remains to be explored. 278

Interestingly, our results suggest that behavioral plasticity may be the ancestral state of 279

antipredator responses of P. perezi to P. clarkii cues. NC populations reduced their activity levels 280

in the presence of fed crayfish at day 60, indicating that P. perezi tadpoles can elicit plastic 281

behavioral defenses towards a predator with which they have had no previous contact. Indeed, 282

NUNES ET AL.

14

reduced activity is a common adaptive defense, which reduces prey detectability and encounter 283

rates with predators (Lima and Dill 1990, Kats and Dill 1998). High activity levels in presence of 284

predators have been linked with lower survival in many tadpole species (e.g. Lawler 1989, Skelly 285

1994), including in P. perezi (Gómez-Mestre and Díaz-Paniagua 2011).While innate recognition 286

of chemical stimuli from native predators is common (reviewed in Wisenden 2003), there are only 287

few such examples involving exotic predators (e.g. Epp and Gabor 2008, Rehage et al. 2009, 288

Bourdeau et al. 2013). Naive populations responding to unknown predators often share an 289

evolutionary history with native predators closely related to the novel predator (Epp and Gabor 290

2008, Rehage et al. 2009). The original geographical range of the white-clawed crayfish 291

Austropotamobius pallipes, the only native crayfish species in Portugal, did not reach Portugal 292

south of the river Tejo basin (Almaça 1991), suggesting that the responses of NC populations are 293

not likely due to phylogenetic similarity to a native predator in the community (Strauss et al. 294

2006). Instead, responses may have been facilitated by the detection of cues from predation events, 295

resulting from injured or digested conspecifics (Ferrari et al. 2010b) or by an innate generalized 296

antipredator response towards predator signals (Gherardi et al. 2011). Native prey can also 297

increase their chances of survival when exposed to new predators by using phenotypically plastic 298

neophobia as an adaptive antipredator strategy (Brown et al. 2013). 299

We also found inducible morphological defenses in response to fed crayfish in one of the LC 300

populations. When exposed to fed crayfish, LC1 tadpoles developed deeper tail fins and muscles, 301

which are common adaptive morphological defenses of larval anurans to predators (e.g. McCollum 302

and Van Buskirk 1996, Relyea 2001, Van Buskirk 2009). Recently, Gómez-Mestre and Díaz-303

Paniagua (2011) showed that similar morphological alterations (deeper tails) - induced by 304

dragonflies - are efficient in decreasing P. perezi mortality imposed by P. clarkii. Predation trials 305

NUNES ET AL.

15

conducted with other anuran species have shown that the tadpoles surviving predation have deeper 306

tails than those sampled before the trials (e.g. Van Buskirk et al. 1997, Van Buskirk and Relyea 307

1998, Teplitsky et al. 2005, Álvarez and Nicieza 2006). These results strongly suggest that the 308

morphological response to P. clarkii in LC1 population is adaptive, indicating that this population 309

has evolved plastic morphological defenses against the invasive crayfish. Previous studies have 310

found high heritability of antipredator responses in tadpoles (Relyea 2005), in particular for 311

activity and tail depth induced by the presence of predators, suggesting that rapid evolutionary 312

response is possible when selection pressure is high. Taken together, our results agree with the 313

idea that invasive species can be a strong selective force generating local adaptation and rapid 314

evolution (Strauss et al. 2006, Sih et al. 2010). While our study reports the fastest evolution of 315

antipredator responses documented for an amphibian (ca. 30 years, 10-15 P. perezi generations), 316

there are several studies reporting rapid evolution in response to selection by invasive predators 317

(Schlaepfer et al. 2005, Freeman and Byers 2006, Strauss et al. 2006, Bourdeau et al. 2013), 318

including in an amphibian (Kiesecker and Blaustein 1997). 319

Our results indicate that LC1 population has developed both constitutive and inducible defenses 320

against P. clarkii. This suggests rapid evolution of a functional integration of constitutive and 321

inducible defenses, promoting a highly effective defensive phenotype, improving adaptation to 322

changing local predation risk and enhancing prey survival (DeWitt et al. 1999, Bourdeau 2012). 323

Theory predicts that high constitutive defenses and moderate/low inducibility - which seems to be 324

the case here - evolve when the frequency of predator encounter rates is relatively high (Poitrineau 325

et al. 2004). Population LC1 is located in a rice field close to the Guadiana River, the river through 326

which P. clarkii likely expanded into Portugal (Ramos and Pereira 1981). Rice fields are one of 327

the preferred habitats of the crayfish in the Iberian Peninsula, where they can attain very high 328

NUNES ET AL.

16

densities (Anastácio et al. 1995). This suggests that population LC1 may have coexisted for a 329

slightly longer time with more dense populations of P. clarkii than, for example, population LC2. 330

In a previous study, Gómez-Mestre and Díaz-Paniagua (2011) did not find morphological 331

responses of P. perezi to P. clarkii in a population from Doñana National Park, where P. clarkii 332

was first found in 1974. While we cannot exclude population differences in intensity of selection 333

or in P. perezi’s potential to evolve antipredator defenses against P. clarkii, the discrepancy 334

between those and the present results could potentially be also explained by the different approach 335

used to examine tadpole morphology in the present study (i.e. geometric morphometric 336

techniques). In any case, studies using multiple populations from a wider geographical area could 337

shed additional light into the role of rapid evolution of antipredator defenses in populations 338

invaded by P. clarkii. 339

We found significant phenotypic divergence in larval morphology among LC/SC and NC 340

populations, the former having deeper headbodies (expanded especially around the gut area) than 341

the latter. While the functional significance of the deeper headbody is unclear at the moment, this 342

result may suggest the existence of a constitutive morphological adaptation in populations 343

coexisting with crayfish, expressed in the form of a larger gut area. Larger guts may allow for a 344

more efficient digestion and rapid growth (Relyea and Auld 2004, Lindgren and Laurila 2005; but 345

see Steiner 2007), which can reduce the time tadpoles are exposed to predators when searching for 346

food, likely reducing the risk of being captured. Actually, a separate experiment showed that P. 347

perezi tadpoles’ tended to grow faster in the presence of fed crayfish than in non-predator 348

environments (Nunes 2012). While this hypothesis agrees well with the generally low activity 349

levels found in LC populations, activity was considerably higher in population SC, suggesting that 350

behavior and gut area were uncoupled in these populations. Alternatively, wider body dimensions 351

NUNES ET AL.

17

may provide protection from gape-limited predators (Kishida and Nishimura 2004; Urban 2007). 352

Although crayfish are not strictly gape-limited predators (Nilsson et al. 2000), they are probably 353

less able to capture and handle prey above a certain size (Renai and Gherardi 2004). Finally, while 354

the populations chosen for this study have similar habitats and experience similar native predator 355

regimes (Ribeiro et al. 2007), we cannot fully exclude the possibility that these differences may 356

result from adaptation to other local habitat features such as competitors or food availability (e.g. 357

Relyea 2002, Van Buskirk and Arioli 2005, Richter-Boix et al. 2010), or from environmental 358

maternal effects (Räsänen and Kruuk 2007). 359

Taking into account the enormous invasion potential of P. clarkii (Capinha and Anastácio 2011), 360

this species is likely to colonize many crayfish-free regions in the future. Given the strong 361

selection pressure imposed by this predator (Cruz and Rebelo 2005, Cruz et al. 2008, Nunes et al. 362

2010, Ficetola et al. 2011), some populations or species may not be able to evolve defensive 363

adaptations rapidly enough to hinder strong declines or even extinctions. Overall, our results 364

suggest that invasive predators may drive rapid evolutionary change in invaded communities. 365

Consequently, examining possible evolutionary responses of prey to invasive predators in invaded 366

communities is crucial for understanding the impact of these on natural populations. However, the 367

alterations in prey traits can be complex, as prey can evolve both constitutive and plastic responses 368

to invasive predators. Considering the variability of the responses and the current rates of spread of 369

exotic predators, our study reinforces the need of multi-population approaches, covering a range of 370

coexistence times, in order to better forecast the long-term impact of biological invasions. 371

372

Acknowledgments 373

NUNES ET AL.

18

We thank Bruno Carreira, Rosário Sumares, João Nunes and Sónia Longueira for their help in the 374

laboratory. We also thank Rui Braz for designing the map of P. perezi populations. Mark Urban 375

and two anonymous reviewers made valuable comments on a previous version of the manuscript. 376

Ángel Ruiz Elizalde kindly provided the P. perezi photograph. The study was conducted in 377

accordance with Portuguese legislation; the collection permit was provided by the Portuguese 378

Instituto da Conservação da Natureza e da Biodiversidade (Licence n.º228/2009/CAPT/TRANS). 379

This research was funded by the FCT Project POCI/BIA-BDE/56100/2004 (to RR), FCT grant 380

SFRH/BD/29068/2006 (ALN) and by Stiftelsen för Zoologisk Forskning (ALN). 381

382

Literature Cited 383

Almaça, C. 1991. L’ecrevisse a pieds blancs, Astacus pallipes Lereboullet 1858, au Portugal. 384

L’Astaciculteur de France 28:11-16. 385

Almeida, N. F., P. Almeida, H, Gonçalves, F. Sequeira, J. Teixeira, and F. Almeida. 2001. Guias 386

FAPAS: Anfíbios e Répteis de Portugal. FAPAS and Câmara Municipal do Porto, Porto, 387

Portugal. 388

Álvarez, D., and A.G. Nicieza. 2006. Factors determining tadpole vulnerability to predators: can 389

prior experience compensate for a suboptimal shape? Evolutionary Ecology 20:523-534. 390

Anastácio, P. M., S. N. Nielsen, J. C. Marques, and S. E. Jorgensen. 1995. Integrated production of 391

crayfish and rice: a management model. Ecological Engineering 4:199-210. 392

Benard, M. F. 2004. Predator-induced phenotypic plasticity in organisms with complex life cycles. 393

Annual Review of Ecology, Evolution and Systematics 35:651-673. 394

NUNES ET AL.

19

Bourdeau, P. E. 2012. Intraspecific trait cospecialization of constitutive and inducible 395

morphological defences in a marine snail from habitats with different predation risk. Journal 396

of Animal Ecology 81:849-858. 397

Bourdeau, P. E., K. L. Pangle, E. M. Reed, and S. D. Peacor. 2013. Finely-tuned response of 398

native prey to an invasive predator in a freshwater system. Ecology 94:1449-1455. 399

Brown, G. E., M. C. O. Ferrari, C. K. Elvidge, I. Ramnarine, and D. P. Chivers. 2013. 400

Phenotypically plastic neophobia: a response to variable predation risk. Proceedings of the 401

Royal Society of London B 280:2012-2712. 402

Capinha, C., and P. Anastácio. 2011. Assessing the environmental requirements of invaders using 403

ensembles of distribution models. Diversity and Distributions 17:13-24. 404

Clark, C. W., and C. D. Harvell. 1992. Inducible defences and the allocation of resources: a 405

minimalist model. The American Naturalist 139:521-539. 406

Cox, J. G., and S. L. Lima. 2006. Naiveté and an aquatic-terrestrial dichotomy in the effects of 407

introduced predators. Trends in Ecology and Evolution 21:674-680. 408

Cruz, M. J., and R. Rebelo. 2005. Vulnerability of southwest Iberian amphibians to an introduced 409

crayfish, Procambarus clarkii. Amphibia-Reptilia 26:293-303. 410

Cruz, M. J., and R. Rebelo. 2007. Colonization of freshwater habitats by an introduced crayfish, 411

Procambarus clarkii, in Southwest Iberian Peninsula. Hydrobiologia 575:191-201. 412

Cruz, M. J., P. Segurado, M. Sousa, and R. Rebelo. 2008. Collapse of the amphibian community 413

of the Paul do Boquilobo Natural Reserve (central Portugal) after the arrival of the exotic 414

American crayfish Procambarus clarkii. Herpetological Journal 18:197-204. 415

DeWitt, T. J., A. Sih, and J. A. Hucko. 1999. Trait compensation and cospecialization in a 416

freshwater snail: size, shape and antipredator behaviour. Animal Behaviour 58:397-407. 417

NUNES ET AL.

20

Edgell, T. C., B. R. Lynch, G. C. Trussell, and A. R. Palmer. 2009 Experimental evidence for the 418

rapid evolution of behavioral canalization in natural populations. American Naturalist 419

174:434-440. 420

Epp, K. J., and C. R. Gabor. 2008. Innate and learned predator recognition mediated by chemical 421

signals in Eurycea nana. Ethology 114:607-615. 422

Ferrari, M. C. O., A. K. Manek, and D. P. Chivers. 2010a. Temporal learning of predation risk by 423

embryonic amphibians. Biology Letters 6:308-310. 424

Ferrari, M. C. O., B. D. Wisenden, and D. P. Chivers. 2010b. Chemical ecology of predator-prey 425

interactions in aquatic ecosystems: a review and prospectus. Canadian Journal of Zoology 426

88:698-724. 427

Ficetola, F. G., M. E. Siesa, R. Manenti, L. Bottoni, F. De Bernardi, and E. Padoa-Schioppa. 2011. 428

Early assessment of the impact of alien species: differential consequences of an invasive 429

crayfish on adult and larval amphibians. Diversity and Distributions 17:1141-1151. 430

Fisk, D. F., L. C. Latta, R. A. Knapp, and M. E. Pfrender. 2007. Rapid evolution in response to 431

introduced predators I: rates and patterns of morphological and life-history trait divergence. 432

BMC Evolutionary Biology 7:22. 433

Freeman, A.S., and J. E. Byers. 2006. Divergent induced responses to an invasive predator in 434

marine mussel populations. Science 313:831-833. 435

Geiger, W., P. Alcorlo, A. Baltanás, and C. Montes. 2005. Impact of an introduced crustacean on 436

the trophic webs of Mediterranean wetlands. Biological Invasions 7:49-73. 437

Gherardi, F. 2006. Crayfish invading Europe: the case study of Procambarus clarkii. Marine and 438

Freshwater Behaviour and Physiology 39:175-191. 439

NUNES ET AL.

21

Gherardi, F., K. M. Mavuti, N. Pacini, E. Tricarico, and D. M. Harper. 2011. The smell of danger: 440

chemical recognition of fish predators by the invasive crayfish Procambarus clarkii. 441

Freshwater Biology 56:1567-1578. 442

Gómez-Mestre, I., and C. Díaz-Paniagua. 2011. Invasive predatory crayfish do not trigger 443

inducible defences in tadpoles. Proceedings of the Royal Society of London B 278:3364-444

3370. 445

Gosner, K. L. 1960. A simplified table for staging anuran embryos and larvae with notes on 446

identification. Herpetologica 16:183-190. 447

Holdich, D. M., J. D. Reynolds, C. Souty-Grosset, and P. J. Sibley. 2009. A review of the ever 448

increasing threat to European crayfish from non-indigenous crayfish species. Knowledge and 449

Management of Aquatic Ecosystems 394-395:11. 450

Johansson, F., B. Lederer, and M. Lind. 2010. Trait performance correlations across life stages 451

under environmental stress conditions in the common frog, Rana temporaria. PLoS One 452

5:e11680. 453

Kats, L. B., and L. M. Dill. 1998. The scent of death: chemosensory assessment of predation risk 454

by prey animals. Ecoscience 5:361-394. 455

Kerby, J. L., S. P. D. Riley, L. B. Kats, and P. Wilson. 2005. Barriers and flow as limiting factors 456

in the spread of an invasive crayfish (Procambarus clarkii) in southern California streams. 457

Biological Conservation 126:402-409. 458

Kiesecker, D. G., and A. R. Blaustein. 1997. Population differences in responses of red-legged 459

frogs (Rana aurora) to introduced bullfrogs (Rana catesbeiana). Ecology 78:1752-1760. 460

Kishida, O., and K. Nishimura. 2004. Bulgy tadpoles: inducible defense morph. Oecologia 461

140:414-421. 462

NUNES ET AL.

22

Lawler, S. P. 1989. Behavioural responses to predators and predation risk in four species of larval 463

anurans. Animal Behaviour 38:1039-1047. 464

Lima, S. L., and L. M. Dill. 1990. Behavioral decisions made under the risk of predation: A review 465

and prospectus. Canadian Journal of Zoology 68:619-640. 466

Lindgren, B., and A. Laurila. 2005. Proximate causes of adaptive growth rates: growth efficiency 467

variation among latitudinal populations of Rana temporaria. Journal of Evolutionary 468

Biology 18:820-828. 469

Losos, J. B., T. W. Schoener, and D. A. Spiller. 2004. Predator-induced behaviour shifts and 470

natural selection in field-experimental lizard populations. Nature 432:505-508. 471

Loureiro, A., N. Ferrand de Almeida, M. A. Carretero, and O. S. Paulo (eds.) 2008. Atlas dos 472

Anfíbios e Répteis de Portugal. Instituto da Conservação da Natureza e da Biodiversidade, 473

Lisboa, Portugal. 474

Magurran, A. E. 1990. The adaptive significance of schooling as an anti-predator defence in fish. 475

Annales Zoologici Fennici 27:3-18. 476

McCollum, S. A., and J. Van Buskirk. 1996. Costs and benefits of a predator-induced polyphenism 477

in the gray treefrog Hyla chrysoscelis. Evolution 50:583-593. 478

McGhee, K. E., L. M. Pintor , E. L.Suhr, and A. M. Bell. 2008. Maternal exposure to predation 479

risk decreases offspring antipredator behavior and survival in threespined stickleback. 480

Functional Ecology 26:932-940. 481

Moran, N. A. 1992. The evolutionary maintenance of alternative phenotypes. American Naturalist 482

139:971-989. 483

NUNES ET AL.

23

Nilsson, P. A., K. Nilsson, and P. Nyström. 2000. Does risk of intraspecific interactions induce 484

shifts in prey-size preference in aquatic predators? Behavioral Ecology and Sociobiology 485

48:268-275. 486

Nunes, A. L. 2012. Predator-prey interactions between an invasive crayfish predator and native 487

anurans: ecological and evolutionary outcomes. PhD thesis, University of Lisbon. 488

Nunes, A. L., M. J. Cruz, M. Tejedo, A. Laurila, and R. Rebelo. 2010. Nonlethal injury caused by 489

an invasive alien predator and its consequences for an anuran tadpole. Basic and Applied 490

Ecology 11:645-654. 491

Nunes, A. L., A. Richter-Boix, A. Laurila, and R. Rebelo. 2013. Do anuran larvae respond 492

behaviourally to chemical cues from an invasive crayfish predator? A community-wide 493

study. Oecologia 171:115-127. 494

Orizaola, G., E. Dahl, and A. Laurila. 2012. Reversibility of predator-induced plasticity and its 495

effect at a life-history switch point. Oikos 121:44-52. 496

Orizaola, G., E. Dahl, A. G. Nicieza, and A. Laurila. 2013. Larval life history and anti-predator 497

strategies are affected by breeding phenology in an amphibian. Oecologia 171:873-881. 498

O'Steen, S., A. J. Cullum, and A. F. Bennet. 2002. Rapid evolution of escape ability in Trinidadian 499

guppies (Poecilia reticulata). Evolution 56:776-784. 500

Padilla, D. K., and S. C. Adolph. 1996. Plastic inducible morphologies are not always adaptive: 501

The importance of time delays in a stochastic environment. Evolutionary Ecology 10:105-502

117. 503

Peluc, S. I., T. S. Sillet, J. T. Rotenberry, and C. K. Ghalambor. 2008. Adaptive phenotypic 504

plasticity in an island songbird exposed to a novel predation risk. Behavioral Ecology 505

19:830-835. 506

NUNES ET AL.

24

Poitrineau, K., S. P. Brown, and M. E. Hochberg. 2004. The joint evolution of defence and 507

inducibility against natural enemies. Journal of Theoretical Biology 231:389-396. 508

Polo-Cavia, N., A. Gonzalo, P. Lopez, and J. Martin. 2010. Predator recognition of native but not 509

invasive turtle predators by naive anuran tadpoles. Animal Behaviour 80:461-466. 510

Ramos, M. A., and T. M. Pereira. 1981. Um novo Astacidae para a fauna portuguesa: 511

Procambarus clarkii (Girard, 1852). Boletim do Instituto Nacional de Investigação das 512

Pescas, Lisboa 6:37-47. 513

Räsänen, K., and L. E. B. Kruuk. 2007. Maternal effects and evolution at ecological time scales. 514

Functional Ecology 21:408-421. 515

Rehage, J. S., K. L. Dunlop, and W. F. Loftu. 2009. Antipredator responses by native mosquitofish 516

to non-native cichlids: an examination of the role of prey naiveté. Ethology 115:1-11. 517

Relyea, R. A. 2001. Morphological and behavioral plasticity of larval anurans in response to 518

different predators. Ecology 82:523-540. 519

Relyea, R. A. 2002. Local population differences in phenotypic plasticity: Predator-induced 520

changes in wood frog tadpoles. Ecological Monographs 72:77-93. 521

Relyea, R. A. 2005. The heritability of inducible defenses in tadpoles. Journal of Evolutionary 522

Biology 18:856-866. 523

Relyea, R. A., and J. R. Auld. 2004. Having the guts to compete: how intestinal plasticity explains 524

costs of inducible defenses. Ecology Letters 7:869-875. 525

Renai, B., and F. Gherardi. 2004. Predatory efficiency of crayfish: comparison between indigenous 526

and non-indigenous species. Biological Invasions 6:89-99. 527

Ribeiro, F., R. Beldade, M. Dix, and J. Bochechas. 2007. Carta Piscícola Nacional Direcção Geral 528

dos Recursos Florestais-Fluviatilis, Lda. Publicação Electrónica (versão 01/2007). 529

NUNES ET AL.

25

Richter-Boix, A., G. A. Llorente, and A. Montori. 2007. A comparative study of predator-induced 530

phenotype in tadpoles across a pond permanency gradient. Hydrobiologia 583:43-56. 531

Richter-Boix, A., C. Teplitsky, B. Rogell, and A. Laurila. 2010. Local selection modifies 532

phenotypic divergence among Rana temporaria populations in the presence of gene flow. 533

Molecular Ecology 19:716-731. 534

Rohlf, F. J. 2007. tpsRelw. Department of Ecology and Evolution. State Univ. New York, Stony 535

Brook. New York, USA. 536

Rohlf, F. J. 2008. tpsDig2. Department of Ecology and Evolution. State Univ. New York, Stony 537

Brook. New York, USA. 538

Rohlf, F. J., and L. F. Marcus. 1993. A revolution in morphometrics. Trends in Ecology and 539

Evolution 8:129-132. 540

Rohlf, F. J., and D. E. Slice. 1990. Extensions of the Procrustes method for the optimal 541

superimposition of landmarks. Systematics Zoology 39:40-59. 542

Salo, P., E. Korpimäki, P. B. Banks, M. Nordström, and C. R. Dickman. 2007. Alien predators are 543

more dangerous than native predators to prey populations. Proceedings of the Royal Society 544

B 274:1237-1243. 545

Schlaepfer, M. A., P. W. Sherman, B. Blossey, and M. C. Runge. 2005. Introduced species as 546

evolutionary traps. Ecology Letters 8:241-246. 547

Sheets, H. D. 2009. MakeFan7. Department of Physics, Canisius College, Buffalo, USA. 548

Sih, A., D. I. Bolnick, B. Luttbeg, J. L. Orrock, S. D. Peacor, L. M. Pintor, E. Preisser, J. S. 549

Rehage, and J. R. Vonesh. 2010. Predator-prey naiveté, antipredator behavior, and the 550

ecology of predator invasions. Oikos 119:610-621. 551

NUNES ET AL.

26

Skelly, D. K. 1994. Activity level and the susceptibility of anuran larvae to predation. Animal 552

Behaviour 47:465-468. 553

Steiner, U. K. 2007. Linking antipredator behaviour, ingestion, gut evacuation and costs of 554

predator-induced responses in tadpoles. Animal Behaviour 74:1473-1479. 555

Strauss, S. Y., J. A. Lau, and S. P. Carroll. 2006. Evolutionary responses of natives to introduced 556

species: what do introductions tell us about natural communities? Ecology Letters 9:354-557

371. 558

Sündermann, D., M. Scheumann, and E. Zimmermann. 2008. Olfactory predator recognition in 559

predator-naive gray mouse lemurs (Microcebus murinus). Journal of Comparative 560

Psychology 122:146-155. 561

Teplitsky, C., S. Plenet, J. - P. Léna, N. Mermet, E. Malet, and P. Joly. 2005. Escape behaviour 562

and ultimate causes of specific induced defences in an anuran tadpole. Journal of 563

Evolutionary Biology 18:180-190. 564

Thompson, J. N. 1998. Rapid evolution as an ecological process. Trends in Ecology and Evolution 565

13:329-332. 566

Tollrian, R., and C. D. Harvell (eds.). 1999. The ecology and evolution of inducible defenses. 567

Princeton University Press, Princeton, USA. 568

Trussell, G. C., and L. D. Smith. 2000. Induced defenses in response to an invading crab predator: 569

An explanation of historical and geographic phenotypic change. Proceedings of the National 570

Academy of Sciences USA 97: 2123-2127. 571

Urban, M. C. 2007. The growth–predation risk trade-off under a growing gape-limited predation 572

threat. Ecology 88:2587-2597. 573

NUNES ET AL.

27

Van Buskirk, J. 2009. Natural variation in morphology of larval amphibians: Phenotypic plasticity 574

in nature? Ecological Monographs 79:681-705. 575

Van Buskirk, J., and R. A. Relyea. 1998. Selection for phenotypic plasticity in Rana sylvatica 576

tadpoles. Biological Journal of the Linnean Society 65:301-328. 577

Van Buskirk, J., and M. Arioli. 2005. Habitat specialization and adaptive phenotypic divergence of 578

anuran populations. Journal of Evolutionary Biology 18:596-608. 579

Wisenden, B. D. 2003. Chemically mediated strategies to counter predation. Pages 236-251 in: S. 580

P. Collin, and N. J. Marshall, editors. Sensory Processing in Aquatic Environments. 581

Springer-Verlag, New York, USA. 582

Zelditch, M. L., D. L. Swiderski, H. D. Sheets, and W. L. Fink. 2004. Geometric morphometrics 583

for biologists: a primer. Academic Press, San Diego, USA. 584

NUNES ET AL.

28

Table 1. Area, coordinates, elevation above sea level of collection sites, type of water body,

approximate coexistence time between Procambarus clarkii and Pelophylax perezi and dates of

collection and of the start of the experiment for each of the five studied populations of P. perezi

Population LC1 LC2 SC NC1 NC2

Area Caia Caia Melides Monchique Monchique

Type of water body

Rice field on

the bank of the

Caia river

Caia river

Melides

stream and

rice fields

Monchique

river

Affluent of

the Seixe

river

Coordinates 38º52’N

7º03’W

39º07’N

7º17’W

38º08’N

8º44’W

37º17’N

8º33’W

37º22’N

8º38’W

Elevation a.s.l 170 m 270 m 25 m 200 m 100 m

Coexistence time 29 years 29 years 19 years 0 years 0 years

Date of collection 10-June 16-June 08-July 18-June 18-June

Number of clutches 8 10 10 9 12

Start of experiment 02-July 02-July 17-August 04-August 05-August

NUNES ET AL.

29

Figure legends

FIG. 1. Map showing the location of the studied Pelophylax perezi populations (LC – Longer

Coexistence, SC – Shorter Coexistence and NC – No Coexistence populations). The black

triangle indicates the location where P. clarkii was first introduced in Spain and the white triangle

indicates the site where crayfish presence was first registered in Portugal. Grey areas indicate

regions >700 m a.s.l.

FIG. 2. Activity rate of Pelophylax perezi tadpoles (mean ± SE) of the five studied populations

raised either in the presence (black square) or absence (white square) of crayfish, measured on

days 10 and 60 of the experiment. LC - Longer Coexistence, SC - Shorter Coexistence and NC -

No Coexistence populations.

FIG. 3. Morphology of Pelophylax perezi tadpoles of the five studied populations raised either in

the presence (black square) or absence (white square) of crayfish, measured after 60 days of the

experiment. LC - Longer Coexistence, SC - Shorter Coexistence and NC - No Coexistence

populations. Values are mean relative warps (RW) scores (± SE) for the first two relative warps

(RW1 and RW2). The figures on the right (grid plots) are thin-plate spline visualizations obtained

from tpsRelw (Rohlf 2007) and illustrate shape variation along the axes, representing the

(positive and negative) extremes of the range of phenotypes found for each RW axis (indicated

by the black arrows).

NUNES ET AL.

30

PLATE 1. Iberian waterfrog (Pelophylax perezi) at a breeding pond. Photo credit: Ángel Ruiz

Elizalde.

NUNES ET AL.

31

Figure 1

NUNES ET AL.

32

Figure 2

0

5

10

15

20

25

30

35

40

Ta

dp

ole

s a

ctive

(%

)

Day of experiment

Population

LC1 LC2 SC NC1 NC2

10 10 10 10 1060 60 60 60 60

NUNES ET AL.

33

Figure 3

No predator Crayfish

Population

-0.04

-0.03

-0.02

-0.01

0.00

0.01

0.02

0.03

0.04

RW

1 =

39

.74

%

LC1 LC2 SC NC1 NC2

a)

-0.02

-0.01

0.00

0.01

0.02

0.03

RW

2 =

15.0

4%

b)

Population

LC1 LC2 SC NC1 NC2

NUNES ET AL.

34

PLATE 1.