Response of predator-naive newt larvae to food and predator presence

Response of predator-naive newt larvae to foodand predator presence

Germán Orizaola and Florentino Braña

Abstract: For naive prey, the ability to recognize predators is advantageous at the time of first predator encounter. Af-ter predator detection, prey could avoid the risk of being eaten by modifying the patterns of activity and use of habitat.The use of refuges is considered one of the most widespread antipredator tactics. In laboratory experiments, we testedthe effects of food (cladocerans, Daphnia sp.) and predator (brown trout, Salmo trutta) presence on larval newt(Triturus helveticus and Triturus alpestris) refuging behaviour, and the effect of refuge use on larval survival. Newt lar-vae came from lakes without fish and were reared in the laboratory from the egg stage so that they lacked any previ-ous experience with brown trout. Larvae of both species increased refuge use and reduced significantly their emergencerate in the presence of predators. Food had no effect on refuge use or activity. For both species, refuge availability in-creased larval survival by more than twofold when the predator was present. These results suggest that naive newt lar-vae are able to detect predators and show an innate antipredator response.

Résumé : Chez les proies inexpérimentées, la capacité de reconnaître les prédateurs constitue un avantage lors de lapremière rencontre avec un prédateur. Une fois le prédateur détecté, la proie peut éviter le risque d’être mangée en mo-difiant ses patterns d’activité et son utilisation de l’habitat. L’utilisation de refuges est reconnue comme l’une des tacti-ques anti-prédateurs les plus répandues. Des expériences de laboratoire nous ont permis de vérifier les effets de laprésence de nourriture (cladocères, Daphnia sp.) et de prédateurs (truite brune, Salmo trutta) sur le comportement dedissimulation de larves de tritons (Triturus helveticus et Triturus alpestris), ainsi que les effets de l’utilisation de refu-ges sur la survie des larves. Les larves proviennent de lacs sans poissons et elles ont été élevées en laboratoire depuisl’oeuf; elles n’ont donc aucune expérience de la présence de la truite brune. En présence de prédateurs, les larves desdeux espèces utilisent plus les refuges et réduisent leur taux d’émergence. La présence de nourriture reste sans effet surl’utilisation des refuges ou sur l’activité. Chez les deux espèces, la disponibilité de refuges accroît la survie des larvesen présence d’un prédateur par un facteur de plus de deux. Ces données indiquent que les larves inexpérimentées detritons sont capables de reconnaître les prédateurs et qu’elles possèdent une réaction anti-prédateurs innée.

[Traduit par la Rédaction] Orizaola and Braña 1850

Introduction

Prey species should be able to detect predator presence todiminish the risk of being eaten. In aquatic ectotherms, pred-ator detection can be based on visual, mechanical, or chemi-cal predator cues or be mediated by alarm or injury signalsemitted by prey (Wilson and Lefcort 1993; Chivers andSmith 1998; Kats and Dill 1998). One crucial point for preysurvival is established at the time of first predator encounterwhen naive prey should be able to react to the predator pres-ence in a way that increases its survival possibilities(Magurran 1990). Several studies have signalled the require-ment of previous predation experience and learning to de-velop antipredator decisions (Magurran 1989; Mathis andSmith 1993; Chivers and Smith 1994, 1995). Some naive or-ganisms can also recognize predation risk without direct ex-

posure by using alarm cues released by injured conspecificsor feeding substances released by predators (Wilson andLefcort 1993; Chivers and Smith 1998; Chivers and Mirza2001). Antipredator responses in naive animals not previ-ously exposed to any predation cue have been reported in afew cases (e.g., for fathead minnows, Phoxinus phoxinus,Magurran 1990; for roach, Rutilus rutilus, Jachner 2001).Previous studies on the role of experience on the develop-ment of amphibian antipredator behaviour have not shownconclusive results: some species modify their defensivestrategies with learning (e.g., Semlitsch and Reyer 1992),whereas others develop protective responses without any ex-perience (e.g., Kats et al. 1998; Mathis et al. 2003).

One of the most widespread behaviours developed by preyto avoid predators is the use of refuges (Sih et al. 1985,1988). An increase in refuge use is associated with reduc-tions of food intake, of mating opportunities, or some physi-ological costs (Lawler 1989; Horat and Semlitsch 1994;Martín and López 1999). Therefore, prey is expected toevolve flexible strategies to manage the use of refuges, mini-mizing the negative costs associated with inactive periods(Martín and López 1999). In amphibians, activity reductionassociated with refuge use may have, in addition to othercosts, a negative effect on growth and developmental rates,increasing the time necessary to reach metamorphosis (Sih

Can. J. Zool. 81: 1845–1850 (2003) doi: 10.1139/Z03-160 © 2003 NRC Canada

1845

Received 11 April 2003. Accepted 11 September 2003.Published on the NRC Research Press Web site athttp://cjz.nrc.ca on 22 December 2003.

G. Orizaola1 and F. Braña. Departamento de Biología deOrganismos y Sistemas, Universidad de Oviedo, c/Catedrático Rodrigo Uría s/n, 33071 Oviedo, Spain.

1Corresponding author (e-mail: [email protected]).

J:\cjz\cjz8111\Z03-160.vpDecember 12, 2003 10:45:34 AM

Color profile: DisabledComposite Default screen

et al. 1988; Jackson and Semlitsch 1993), and therefore in-creasing the time that larvae are exposed to aquatic preda-tors and to the risk of habitat desiccation (Laurila andKujasalo 1999). The development of this kind of behav-ioural antipredator responses has been well studied in anuranlarvae (see Lawler 1989; Alford 1999; Van Buskirk 2000),but less so in the case of tailed amphibians (Holomuzki1986; Semlitsch 1987; Sih et al. 1992; Van Buskirk andSchmidt 2000).

The aim of this paper was to investigate the capability ofpredator-naive newt larvae to alter their behaviour under adisturbing situation associated with predator presence. In thefirst experiment, we used larvae of two newt species, pal-mate newt (Triturus helveticus Razoumowski, 1789) and al-pine newt (Triturus alpestris Laurenti, 1768), and tested theeffect of food (Daphnia sp.) and predator presence(nonlethal brown trout, Salmo trutta) on overall refuge use(time in refuges and refuge emergence rate), activity, andfood consumption. In a second experiment, we quantifiedthe differences in survival when larvae were exposed to le-thal predatory fish in the presence or absence of refuges.

Materials and methods

Study animalsGravid T. alpestris (n = 31) and T. helveticus (n = 13) fe-

males were collected in June and July 2000 from two smallmountain lakes of Asturias, northern Spain (Veneros and LaSobia, elevation 1300 and 1200 m, respectively), that haveno fish. Females were allowed to lay eggs in natural vegeta-tion arranged in laboratory aquaria and then returned to thelake of origin. At hatching, each newt larva was introducedinto an individual 50-mL plastic flask. The larvae were heldat constant temperature (17 °C) at a 12 h light : 12 h darkphotoperiod and were fed live zooplankton (cladocerans andcopepods) and frozen chironomid (Chironomidae) larvae ev-ery 2 days. Only larvae with well-developed limbs wereused in the experiment. Larvae were measured 2 days beforethe experiments and assigned to each treatment to avoid sizedifferences among groups. Total length (mean ± SE) of thelarvae used in the trials was 16.94 ± 0.11 mm for T. hel-veticus and 17.80 ± 0.09 mm for T. alpestris. To standardizehunger level, the newt larvae were starved for 2 days beforethe experiments. After the experiments, all the larvae werereleased to their places of origin. Brown trout predatorscame from the spawning of wild parents from the ColorRiver, Asturias (northern Spain). Brown trout is the mostcommon fish species in rivers of the area (Reyes-Gavilán etal. 1995) and is known to be an active predator of newts(Braña et al. 1996). Brown trout used in the trials (forklength 116.80 ± 3.90 mm, mean ± SE) were starved for3 days before the experiments to minimize the effect of sati-ation on predator performance and to avoid faecal contami-nation of the water. Between trials, brown trout were fedcommercial pellets and chironomid larvae. The experimentswere carried out in November and December 2000, in aroom held at constant 17 °C.

Experimental proceduresThe effect of food and predator presence on newt behav-

iour was studied in white plastic containers (40 cm long,

31 cm wide), which were filled with about 9 L of dechlor-inated tap water to a depth of 7.5 cm. A transparent acetatesheet divided the container into two compartments, a smallerone (12 cm × 31 cm) to lodge the brown trout predator and alarger one (28 cm × 31 cm) for the prey. The partition wasperforated to allow water diffusion but not animal move-ment between compartments. The bottom of the prey com-partment was marked with lines dividing it into fourquadrants. To provide refuges to the newts, we placed in thecentre of each quadrant a piece of mesh (8 cm × 8 cm) ele-vated 1 cm above the bottom by four polystyrene cubes onthe corners of the mesh. The water was aerated continuouslyby a single airstone located in the predator compartment.Two predator treatments (one brown trout or no predator)and two food treatments (15 Daphnia sp. added to the newtcompartment or no food) were tested. A total of 14 browntrout were used in the trials, introducing one brown trout inthe container 5 min before introducing the newt to allow thefish to acclimate and to allow the predator odour to diffusethroughout the container. Then a newt larva was transferredto the centre of the prey compartment and was left to accli-mate for 5 min. After the acclimation period, the trial startedand we recorded the position of the larvae every minute for30 min. From these recordings, we calculated the percentageof time that the larvae spent under refuge, the number oftimes that the larvae left the refuge (for larvae that use therefuge at least once), and larval activity (the number oftimes that larvae changed quadrants). The number of Daph-nia sp. remaining at the end of the trial was also recordedfor the food treatments. Between trials the containers werethoroughly rinsed. In total 68 T. alpestris and 56T. helveticus larvae were used in this experiment. Therewere no within-species differences in larval length amongthe four treatment combinations (ANOVA; T. helveticus:F[3,63] = 0.474, P = 0.701; T. alpestris: F[3,39] = 0.175, P =0.912).

We studied the effect of refuge availability (four refugesor no refuge) on the survival of the two newt species inwhite plastic containers (60 cm long, 40 cm wide). A remov-able plastic mesh partition divided the containers into twocompartments, a smaller one to hold the predator and alarger one where pieces of plastic similar to those used inthe previous experiment served as refuges for the newts. Thecontainers were filled to a depth of 7.5 cm with 20 L ofdechlorinated tap water that was continuously aerated withone airstone. The order of the trials was completely random-ized, rinsing the containers after each trial. One brown troutwas used as the predator and it was allowed to acclimate for5 min before introducing the newts. Because only ninebrown trout were available, they were evenly assigned to thedifferent trials. Time between trials was at least 12 h, and sa-tiation problems were not detected. Because of the timeelapsed between the trials and the small size of prey avail-able, it is unlikely that feeding cues emitted by brown troutcould have had any effect on newt larval behaviour. Fiveminutes before a trial started, three newt larvae were trans-ferred to the centre of the larger compartment and the trialstarted when the partition was removed. Because of the dif-ferent availability of individuals from the two species, treat-ments involving T. alpestris were replicated 11 times,whereas treatments with T. helveticus were replicated 9 times.

© 2003 NRC Canada

1846 Can. J. Zool. Vol. 81, 2003



There were no within-species differences in larval bodylength among treatments (ANOVA; T. helveticus: F[1,41] =0.025, P = 0.873; T. alpestris: F[1,64] = 0.337, P = 0.563) oramong the groups of three larvae in each treatment (ANOVA;T. helveticus: F[17,36] = 1.110, P = 0.381; T. alpestris:F[17,25] = 0.433, P = 0.961).

Both newt species are abundant and not endangered in thestudy area, inhabiting the majority of ponds and lakes. In ad-dition, special considerations have been given to reduce thepain of prey animals. The first experiment involved no phys-ical contact between predators and prey, whereas in the sec-ond experiment, when prey consumption was necessary toaccomplish the objectives of the study, trials were performedusing the minimum number of replicates.

Statistical analysesWe used a three-way ANOVA (newt species, predator

presence, and food as factors) to test for differences in thepercentage of time spent in refuge and number of refugeemergences, and a three-way ANCOVA (with the number oftimes that each larva was outside the refuge as covariate) toanalyse larval activity. A two-way ANOVA was used to ex-amine the effect of predator presence on the number ofDaphnia sp. eaten by the two newt species. Differences inlarval survival in relation to refuge availability were ana-lysed using a G test. The level of significance was set at α =0.05 for all tests. Percent data were arcsine square-roottransformed, and the square-root transformation was used inthe case of the number of Daphnia sp. eaten.

Results

Triturus helveticus used the refuges more than T. alpestris(F[1,116] = 40.427, P < 0.001) and both species increased ref-uge use with predator presence (F[1,116] = 13.036, P <0.001), but no effect of food was detected (F[1,116] = 0.057,P = 0.812; Fig. 1). Refuge use increased during the experi-ment for both species. At the end of the trials (30 min), the

percentage of T. helveticus under refuge in the treatmentwith no predator was 57% lower than in treatments with apredator, and it was 23% lower in the case of T. alpestris. Asignificant interaction was detected between newt speciesand predator presence (F[1,116] = 40.427, P < 0.001), reveal-ing a stronger response in T. helveticus larvae than inT. alpestris larvae. Newt larvae not exposed to predatorsmoved more actively in and out of the refuge (F[1,97] =10.169, P = 0.002; Fig. 2). Food had no effect on the num-ber of refuge emergence (F[1,97] = 0.0006, P = 0.981).Triturus alpestris showed a higher rate of refuge emergencethan T. helveticus (F[1,97] = 8.275, P = 0.005). Neither foodnor predator presence had a significant effect on the activityof newt larvae outside refuges (predator: F[1,110] = 2.238,P = 0.137; food: F[1,110] = 2.369, P = 0.127). There was,however, a significant interaction effect between predatorand food (F[1,110] = 8.617, P = 0.004) that was related tohigher activity in trials with food and no predators (P =0.033). The number of Daphnia sp. consumed by newt lar-vae did not differ between species (F[1,120] = 0.156, P =0.693) or predator treatments (F[1,120] = 0.054, P = 0.815).

When the newts were exposed to a lethal predator, larvalsurvival was positively influenced by the presence of refuges(G = 4.37, P < 0.05 for T. alpestris; G = 5.384, P < 0.025for T. helveticus), which were highly used throughout the tri-als (mean percentage of larvae under refuge was 79% forT. alpestris and 96% for T. helveticus). The number of newtlarvae surviving when refuges were available was, on aver-age, 44% higher for T. helveticus and 33% for T. alpestris(Fig. 3).

Discussion

Predator recognition has an obvious associated survivalvalue for prey, and the earlier the recognition is developed,the greater is the benefit. The results of this study show thecapacity of naive newt larvae to respond to predatory fishpresence; the presence of brown trout increases the use of

© 2003 NRC Canada

Orizaola and Braña 1847

Fig. 1. Refuge use expressed as the percentage of time (mean +SE) that newt larvae spent under refuge in relation to predatorpresence and food availability. Treatments were replicated 14times for Triturus helveticus and 17 times for Triturus alpestris.

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.25

Food No food Food No food

Me

an

refu

ge

em

erg

en

ce

s

Predator

No Predator

T. helveticus T. alpestris

Fig. 2. Effect of food and predator presence on the number(mean + SE) of emergences from refuges for newt larvae thatused refuge at least once. Treatments were replicated 14 timesfor T. helveticus and 17 times for T. alpestris.



refuges and reduces the emergence rates of naive newt lar-vae. In the experiments, we used laboratory-reared newt lar-vae from lakes without fish and exposed them to brown troutthat had not previously eaten newts. Also, the newt larvaewere maintained individually isolated during their develop-ment and were individually exposed to predators so that re-sults were free from the effects of previous experience orfrom interactive effects produced by alarm or feeding cuesreleased by conspecifics. Therefore, the results of our exper-iment suggest the existence of innate defensive behaviour.These responses could be fine-tuned with learning in situa-tions of coexistence between predators and prey (Magurran1990; Semlitsch and Reyer 1992; Chivers and Smith 1994).

The observed increase in refuge use and the consequentreduction in activity have been described for different preyspecies as typical responses to predator presence (Sih et al.1988, 1992; Lima 1998; Van Buskirk and Schmidt 2000).An increase in refuge use implies a reduction in the timethat the animal is exposed to predators and, for this reason, ithas been associated with an increase in survival (Sih et al.1988; Lawler 1989; Skelly 1994). In this study, refuge avail-ability increased newt survival by more than twofold inrelation to trials without refuges. Refuge presence also con-tributed to increased habitat complexity, which could reducethe predator–prey encounter rate, especially in visually ori-ented predators such as brown trout, and could also impairpredator visual capabilities and movements (Werner et al.1983; Sredl and Collins 1992; Babbitt and Jordan 1996;Babbitt and Tanner 1997). However, refuge use also reducesthe opportunities for social interactions or foraging (Sih1987; Martín and López 2000). Therefore, prey should beable to evaluate predator persistence to reinitiate normal ac-tivity (Martín and López 1999, 2000). In this experiment, wehave shown that newt larvae under a risky situation had lowemergence rates from refuges, whereas larvae in preda-tor-free treatments moved frequently in and out of the ref-uges. Permanence in refuges could be associated with

periods of prey latency after detecting predator cues or withprey detecting the persistence of risk (Sih 1992).

Defensive behaviour of newt larvae exposed to browntrout could not be strictly considered a specific antipredatorbehaviour but could be interpreted as a response to a generaldisturbance or a conservative response to a potential preda-tor (see Mathis et al. 2003). Additional experiments arenecessary to address such specificity. In any case, the devel-opment of protective behaviours could be advantageous forprey exposed to a wide range of potential predators, as oc-curs in the breeding habitats of newts (Van Buskirk andSchmidt 2000; G. Orizaola, personal observations). Behav-ioural responses such as reduction in activity and increase inrefuge use have been reported for many amphibian speciesexposed to predation risk (Sih et al. 1988, 1992; Kats et al.1998; Van Buskirk and Schmidt 2000; Teplitsky et al. 2003).On the other hand, several studies have shown that differentamphibian species (e.g., Rana sylvatica, Rana temporaria,T. alpestris, T. helveticus) exposed to different predators arealso able to develop similar morphological modifications(deep tails, small bodies) that increase their capacity toescape from predators (Van Buskirk and Relyea 1998;Van Buskirk and Schmidt 2000; Relyea 2001; Van Buskirk2002).

Our results showed that newts recognize risky situationsand display innate responses, and such capacity should fa-vour the maintenance of fish–newt communities in habitatswhere they meet in natural conditions. However, in our studyarea, a definite decrease in newt abundance has been re-ported in fish-inhabited lakes (Braña et al. 1996; G. Orizaolaand F. Braña, unpublished data). The responses observed inthis study were stronger for T. helveticus larvae than forT. alpestris larvae, paralleling the results obtained by VanBuskirk and Schmidt (2000) for the same two species ex-posed to dragonfly (Aeshna cyanea) predators. Several stud-ies have reported that the costs associated with avoidingpredators (reduced activity and increased refuge use) couldbe as important as the direct effects of predation and mightaffect population viability (Abrams 1991; McPeek andPeckarsky 1998; Tyler et al. 1998; Peacor and Werner 2001).Therefore, it is possible that those indirect effects, as well asdirect predation on adults and larvae, could be contributingto the reduction of newt populations in our study area (Brañaet al. 1996).

Acknowledgements

We are especially grateful to Carlos Rodríguez del Vallefor his assistance and friendship. Felipe G. Reyes-Gavilánimproved the manuscript with his suggestions and Eng-lish-style advice. All the animals utilized in this study wereused under collection and experimental permissions pro-vided by the regional authorities (Dirección General deRecursos Naturales y Protección Ambiental, Principado deAsturias). A Ministry of Science and Technology researchproject (BOOS2000-0452) supported this research. A Span-ish Ministry of Education and Culture doctoral fellowshipand a University of Oviedo grant provided financial supportto G. Orizaola.

© 2003 NRC Canada

1848 Can. J. Zool. Vol. 81, 2003

Fig. 3. Effect of refuge presence on T. helveticus and T. alpestrissurvival. Data are expressed as the percentage of surviving larvae(mean + SE) at the end of the experimental period (30 min). Foreach treatment, sample size (n) is 27 for T. helveticus and 33 forT. alpestris.



References

Abrams, P.A. 1991. Life history and the relationship between foodavailability and foraging effort. Ecology, 72: 1242–1252.

Alford, R.A. 1999. Ecology: resource use, competition, and preda-tion. In Tadpoles: the biology of anuran larvae. Edited by R.W.McDiarmid and R. Altig. The University of Chicago Press, Chi-cago, Ill. pp. 240–278.

Babbitt, K.J., and Jordan, F. 1996. Predation on Bufo terrestris tad-poles: effects of cover and predator identity. Copeia, 1996:485–488.

Babbitt, K.J., and Tanner, G.W. 1997. Effects of cover and predatoridentity on predation of Hyla squirella tadpoles. J. Herpetol. 31:128–130.

Braña, F., Frechilla, L., and Orizaola, G. 1996. Effect of introducedfish on amphibian assemblages in mountain lakes of northernSpain. Herpetol. J. 6: 145–148.

Chivers, D.P., and Mirza, R.S. 2001. Importance of predator dietcues in responses of larval wood frogs to fish and invertebratepredators. J. Chem. Ecol. 27: 45–51.

Chivers, D.P., and Smith, R.J.F. 1994. The role of experience andchemical alarm signalling in predator recognition by fatheadminnows, Pimephales promelas. J. Fish Biol. 44: 273–285.

Chivers, D.P., and Smith, R.J.F. 1995. Free-living minnows rapidlylearn to recognize pike as predators. J. Fish Biol. 46: 949–954.

Chivers, D.P., and Smith, R.J.F. 1998. Chemical alarm signalling inaquatic predator–prey systems: a review and prospectus.Ecoscience, 5: 338–352.

Holomuzki, J. 1986. Predator avoidance and diel patterns of micro-habitat use by larval tiger salamanders. Ecology, 67: 737–748.

Horat, P., and Semlitsch, R.D. 1994. Effects of predation risk andhunger of the behaviour of two species of tadpoles. Behav. Ecol.Sociobiol. 34: 393–401.

Jachner, A. 2001. Anti-predator behaviour of naïve compared withexperienced juvenile roach. J. Fish Biol. 59: 1313–1322.

Jackson, M.E., and Semlitsch, R.D. 1993. Paedomorphosis in thesalamander Ambystoma talpoideum: effects of a fish predator.Ecology, 74: 342–350.

Kats, L.B., and Dill, L.M. 1998. The scent of death: chemosensoryassessment of predation risk by prey animals. Ecoscience, 5:361–363.

Kats, L.B., Petranka, J.W., and Sih, A. 1988. Antipredator defensesand the persistence of amphibian larvae with fishes. Ecology,69: 1865–1870.

Laurila, A., and Kujasalo, J. 1999. Habitat duration, predation riskand phenotypic plasticity in common frog (Rana temporaria). J.Anim. Ecol. 68: 1123–1132.

Lawler, S.P. 1989. Behavioural responses to predators and preda-tion risk in four species of larval anurans. Anim. Behav. 38:1039–1047.

Lima, S.L. 1998. Stress and decision making under the risk of pre-dation: recent developments from behavioral, reproductive, andecological perspectives. In Advances in the study of behaviour.Vol. 27. Stress and behaviour. Edited by A.P. Møller, M.Milinski, and P. Slater. Academic Press, London, U.K.pp. 215–290.

Magurran, A.E. 1989. Acquired recognition of predator odour inthe European minnow (Phoxinus phoxinus). Ethology, 82:216–223.

Magurran, A.E. 1990. The inheritance and development of minnowanti-predator behaviour. Anim. Behav. 39: 834–842.

Martín, J., and López, P. 1999. When to come out from a refuge:risk-sensitive and state-dependent decisions in an alpine lizard.Behav. Ecol. 10: 487–492.

Martín, J., and López, P. 2000. Cost of refuge use affect escape de-cisions of iberian rock lizards Lacerta monticola. Ethology, 106:483–492.

Mathis, A., and Smith, R.J.F. 1993. Fathead minnows, Pimephalespromelas, learn to recognize northern pike, Esox lucius, as pred-ators on the basis of chemical stimuli from minnows in thepike’s diet. Anim. Behav. 46: 645–656.

Mathis, A., Murray, K.L., and Hickman, C.R. 2003. Do experienceand body size play a role in responses of larval ringed salaman-ders, Ambystoma annulatum, to predator kairomones? Labora-tory and field assays. Ethology, 109: 159–170.

McPeek, M.A., and Peckarsky, B.L. 1998. Life histories and thestrengths of species interactions: combining mortality, growth,and fecundity effects. Ecology, 79: 867–879.

Peacor, S.D., and Werner, E.E. 2001. The contribution oftrait-mediated indirect effects to the net effects of a predator.Proc. Natl. Acad. Sci. U.S.A. 98: 3904–3908.

Relyea, R.A. 2001. The relationship between predation risk andantipredator responses in larval anurans. Ecology, 82: 541–554.

Reyes-Gavilán, F.G., Garrido, R., Nicieza, A.G., Toledo, M.M.,and Braña, F. 1995. Variability in growth, density and age struc-ture of brown trout populations under contrasting environmentaland managerial conditions. In The ecological basis for rivermanagement. Edited by E.D. Harper and A.J.D. Ferguson. JohnWiley & Sons Ltd., Chichester, U.K. pp. 389–403.

Semlitsch, R.D. 1987. Interactions between fish and salamanderlarvae: costs of predator avoidance or competition? Oecologia,72: 481–486.

Semlitsch, R.D., and Reyer, H.U. 1992. Modifications ofanti-predator behaviour in tadpoles by environmental condition-ing. J. Anim. Ecol. 61: 353–360.

Sih, A. 1987. Predators and prey lifestyles: An evolutionary andecological overview. In Predation: direct and indirect impacts onaquatic communities. Edited by W.C. Kerfoot and A. Sih. Uni-versity Press of New England, Hanover, N.H. pp. 203–224.

Sih, A. 1992. Prey uncertainty and the balancing of antipredatorand feeding needs. Am. Nat. 139: 1052–1069.

Sih, A., Crowley, P., McPeek, M., Petranka, J.W., and Strohmeier,K. 1985. Predation, competition, and prey communities. Annu.Rev. Ecol. Syst. 16: 269–311.

Sih, A., Petranka, J.W., and Kats, L.B. 1988. The dynamics of preyrefuge use: a model and tests with sunfish and salamander lar-vae. Am. Nat. 132: 463–483.

Sih, A., Kats, L.B., and Moore, R.D. 1992. Effects of predatorysunfish on the density, drift, and refuge use of stream salaman-der larvae. Ecology, 73: 1418–1430.

Skelly, D.K. 1994. Activity level and the susceptibility of anuranlarvae to predation. Anim. Behav. 47: 465–468.

Sredl, M.J., and Collins, J.P. 1992. The interaction of predation,competition, and habitat complexity in structuring an amphibiancommunity. Copeia, 1992: 607–614.

Teplitsky, C., Plénet. S., and Joly, P. 2003. Tadpoles’ responses torisk of fish introduction. Oecologia, 134: 270–277.

Tyler, T.J., Liss, W.J., Hoffman, R.L., and Ganio, L.M. 1998. Ex-perimental analysis of trout effects on survival, growth, andhabitat use of two species of Ambystomatid salamanders. J.Herpetol. 32: 345–349.

Van Buskirk, J. 2000. The costs of an inducible defense in anuranlarvae. Ecology, 81: 2813–2821.

Van Buskirk, J. 2002. Phenotypic lability and the evolution ofpredator-induced plasticity in tadpoles. Evolution, 56: 361–370.

Van Buskirk J., and Relyea, R.A. 1998. Selection for phenotypicplasticity in Rana sylvatica tadpoles. Biol. J. Linn. Soc. 65:301–328.

© 2003 NRC Canada

Orizaola and Braña 1849



Van Buskirk J., and Schmidt, B.R. 2000. Predator-induced pheno-typic plasticity in larval newts: trade-offs, selection, and varia-tion in nature. Ecology, 81: 3009–3028.

Werner, E.E., Gilliam, J.F., Hall, D., and Mittelbach, G.G. 1983.An experimental test of the effects of predation risk on habitatuse in fish. Ecology, 64: 1540–1548.

Wilson, D.J., and Lefcort, H. 1993. The effect of predator diet onthe alarm response of red-legged frog, Rana aurora, tadpoles.Anim. Behav. 46: 1017–1019.

© 2003 NRC Canada

1850 Can. J. Zool. Vol. 81, 2003



Documents

Response of predator-naive newt larvae to food and predator presence