Abstract
Interference behaviour (aggression and cannibalism) can be influenced by both intrinsic factors, such as animal physiology, size, or motivation, and extrinsic factors, such as presence of competitors, predators, or prey. Our experiment examined the effects of differences in diet quality of focal salamanders and their opponents, and levels of handling on biting by pairs of larval ringed salamanders, Ambystoma annulatum, before and after presentation of prey. Diet quality and handling affect the intrinsic qualities of body size/condition and stress, respectively. Presence of prey and diet quality of opponents are extrinsic factors. Unstressed larvae on high-quality diets bit their high-condition opponents more frequently than those that had been on low-quality diets in both the presence and absence of food. Stressed larvae (all on high quality diets) showed low levels of biting when food was absent. However, when food was present, the level of biting depended on the diet quality of their opponents: stressed larvae bit opponents on low-quality diets more than opponents on high-quality diets. Overall, both intrinsic and extrinsic factors influenced interference behaviour, with larvae on high-quality diets exhibiting relatively high levels of biting unless they had experienced handling stress, and the effect of handling stress depending on the condition of the opponent in paired interactions.
Introduction
For species with gape-limited predators, selection for rapid-growth can lead to intense competition for food resources, particularly amongst juveniles (Sogard, 1997; Urban, 2008). Variation in competitive success can contribute to variable growth rates, resulting in a large range of body sizes even amongst individuals in the same age class (e.g., Brunkow and Collins, 1998; Echaubard et al., 2010). Size (and associated condition) can then have consequences for the probability of success in competitive encounters (Smith, 1990; Reques and Tejedo, 1996; Doyle, Nolan and Whiteman, 2010).
In many vertebrates, competition is manifested as interference behaviour, which ranges from advertisement displays to overt aggression (biting, wrestling) and cannibalism (Fox, 1975). Levels of interference behaviour can be influenced by intrinsic and/or extrinsic factors. For example, either absolute or relative body size is a common intrinsic variable that influences aggressive behaviour, with larger individuals frequently being the more aggressive (e.g., fish: Prenter, Taylor and Elwood, 2008; salamanders: Mathis and Britzke, 1999; lizards: Sacchi et al., 2009). Extrinsically, levels of aggression can also be influenced by acute or long-term stress, with stress associated with either increases (Mineur et al., 2003) or decreases (Martel and Dill, 1993; Gammie and Stevenson, 2006) in aggressive behaviour. In addition, when resource availability varies either spatially, temporally, or both, aggression can be influenced by whether the resource is present. For example, aggression in male swordtails, Xiphophorus helleri, is higher when both food and females are present than when they are absent (Magellan and Kaiser, 2010).
For amphibians with complex life cycles (aquatic larvae metamorphosing into terrestrial adults), competition for food is particularly intense during the larval stage; individuals that grow quickly experience greater survival, particularly with respect to gape-limited predation (Kishida et al., 2015). Interference competition is particularly important for salamanders, whose larvae, unlike frog tadpoles, have teeth and mouthparts suitable for biting. Both aggressive behaviour and cannibalism are common in larval salamanders (e.g., Walls and Jaeger, 1987; Petranka, 1989; Wildy et al., 2001). For many species of pond-breeding salamanders of the genus Ambystoma, growth has been shown to be density dependent (Stenhouse, Hairston and Cobey, 1983; Petranka, 1989; Wildy et al., 2001), and size is highly variable within a cohort (Stenhouse, Hairston and Cobey, 1983; Jefferson et al., 2014).
Our study examined the effects of an individual’s diet quality, its opponent’s diet quality, extent of handling, and the presence of food on interference behaviour in larval ringed salamanders, Ambystoma annulatum. Manipulation of diet quality should influence the intrinsic factors of body condition and/or size, and different levels of handling should influence intrinsic stress levels. The presence or absence of food and the opponent’s diet quality are extrinsic factors.
Adult ringed salamanders are terrestrial, inhabiting forested areas in the Ozark Plateau in Missouri, Arkansas, and Oklahoma, where they are endemic (Johnson, 2000). Eggs are laid in ephemeral ponds in the early fall, and larvae overwinter in the ponds and metamorphose in the late spring, providing prolonged opportunities for growth. Mortality is very high during this period, with less than 1% of eggs surviving to metamorphosis (Peterson et al., 1991). Because of its limited range, this species is listed as being of conservation concern in the state of Missouri. Therefore, an understanding of how various stressors influence its behaviour are particularly important.
We manipulated diet quality and stress levels of lab-reared larvae and examined effects on interference behaviour (biting) both before and after presentation of food. For unstressed larvae, we tested whether the diet quality of the focal larvae affected level of biting; for stressed larvae, we tested whether the diet quality of the opponent affected level of biting. In general, we predicted that focal larvae would be less aggressive when in poorer condition (on low-quality diets), when in the high stress condition (more handling), and when food is absent.
Handling stress has been found to effect some salamander behaviours (activity), but not others (courtship and mating) (Desmognathus: Ricciardella et al., 2010; Woodley and Lacey, 2010; Bliley and Woodley, 2012; Plethodon: Wack, Ratay and Woodley, 2013; adult Ambystoma: Woodley and Porter, 2016). Our stress treatment also included introduction of the handled salamander into a chamber already occupied by another salamander. In some terrestrial salamanders, individuals that are introduced into another salamander’s home chamber are considered to be territorial intruders, and differences in aggressive behaviour are interpreted with respect to predictions of territoriality (e.g., resident advantage) (e.g., Cupp, 1980; Mathis, Schmidt and Medley, 2000). It is unlikely that this theoretical framework applies to our data because lack of a defensible (patchy) resource and a stable base for territorial (pheromonal) markers makes territoriality unlikely for larval Ambystoma (Mathis et al., 1995).
We tested whether the following conditions affected the level of biting by larvae both before and after presentation of food: for unstressed larvae, we tested whether the diet condition of the focal larva affected level of biting; for stressed larvae, we tested whether the diet quality of the opponent affected level of biting (table 1). Data for stressed and unstressed salamanders were examined separately.
Experimental design for tests of the influence of diet quality, handling stress and the presence of food on interference behavior by pairs of larval ringed salamanders. The design is unbalanced due to a limited number of larvae. For animals that did not experience handling stress (I), all opponents were on high-quality diets, and diet-quality of focal larvae was manipulated. For animals that experienced handling stress (II), all focal larvae were on high-quality diets and diet quality of their opponents was manipulated. For all trials, data on interference behavior (biting) was collected before and after presentation of food. Data were analyzed with separate two-way ANOVAs with repeated measures for stressed and unstressed larvae. Final sample sizes (n) reflect mortality (1 larva on a high-quality diet and 4 larvae on low-quality diets died during the study).
Materials and methods
To minimize the probability that all larvae would be related, we collected approximately five separate clusters of A. annulatum eggs from Bull Shoals Field Station, Taney County, Missouri, in September 2011. After hatching (October), larvae were housed at 23.5 (± 2)°C in the laboratory in individual plastic containers (8.3 × 9.3 × 9.3 cm) filled with a 50:50 dechlorinated tap: pond water mixture; each container had a small feeding dish affixed to one wall 2.5 cm from the bottom. Larvae were kept on a 12:12 light:dark cycle and were fed Daphnia sp. daily until assignment of diet treatments, approximately six weeks after hatching.
Individuals were randomly assigned to unstressed or stressed treatments. Beginning 80 days before testing, half of the unstressed salamanders were randomly assigned to a high-quality diet () and the other half were assigned to a low-quality diet (), while all of the stressed individuals () were fed a high-quality diet. Salamanders were fed 0.1 ml of their respective diets (high-quality: 2 live bloodworms, Tubifex sp., plus 2 frozen brine shrimp, Artemia sp.; low-quality: 6 frozen brine shrimp) every 2 days, with food added to the feeding dish. Our bomb calorimetry of 0.10-0.23 ml samples of brine shrimp () and bloodworms () revealed a statistically significant, difference in total caloric content between the two diets (worm = 0.0165 ± 0.00023 kcal/0.1 ml and brine shrimp = 0.0107 ± 0.00006 kcal/0.1 ml; , ). In addition, digestible content should differ because the exoskeletons of brine shrimp contain chitin (Richards, 1951), which is difficult for amphibians to digest (Simon and Toft, 1991). The brine shrimp that supplemented the high-quality (worm) diets ensured that the high-quality individuals would be familiar with this prey during testing and also added additional caloric content to the high-quality diet.
To confirm that the diets differed in quality, we measured effects of diet on growth and body condition. We measured total length (end of snout to tip of tail) and volume (via volumetric displacement of water in a graduated cylinder) of larvae at the end of the 80-day rearing period, and compared the lengths of salamanders in high-quality and low-quality diets with a two sample t-test. We then performed a length/volume regression and tested for differences in body condition by comparing the residuals for the high-quality diet and low-quality diet salamanders using a two sample t-test (one-tailed).
Each behavioural trial consisted of observations of biting behaviour by a pair of salamanders, with one randomly selected individual being tested in its home chamber without disturbance (hereafter, “unstressed”) and the other removed from its home chamber and placed in the chamber of the other larva (handling stress; hereafter, “stressed”). Because of a limited number of available salamanders (due to the conservation status of this species), the treatments for diet and opponent condition were unbalanced; that is, not all possible treatment combinations are represented (table 1). To minimise possible cumulative effects of low-quality diet and stress, all stressed larvae were fed a high-quality diet and are hereafter referred to as “high condition”. Therefore, all unstressed larvae had high-condition stressed opponents whereas stressed larvae had unstressed opponents, half of which were high condition and half were low condition. Data for stressed and unstressed salamanders were examined separately.
Behavioural tests were conducted at 23.5 (± 2)°C between 10:00 and 18:30 h in February 2012, 2 d after the last feeding to ensure consistent hunger levels. Unstressed individuals (on either high- or low-quality diets) were tested in their home containers and were not handled prior to testing. Stressed individuals were removed from their home containers with a net and placed in the home chambers of their randomly-selected unstressed opponents. The random assignment of individuals resulted in no difference between diet treatment groups in size differences between pairs (mean ± SE: high-condition vs. high-condition trials, 2.7 mm ± 0.43; high-condition vs. low-condition trials, 2.7 mm ± 0.55). Immediately following introduction of the stressed salamander, we recorded the number of bites by both individuals in each pair for 5 min. We then added 5 frozen brine shrimp to the feeding dish and recorded the number of bites for another 5 min.
The effect of diet quality on body size/condition was analyzed with a t-test; these data met the assumptions of parametric statistics. The behavioural data were not normally distributed, so analyses were conducted on aligned-rank transformed data (Higgins and Tantoush, 1994). Because treatments were not balanced we conducted separate 2-way ANOVAs with repeated measures for stressed and unstressed salamanders. For unstressed salamanders, the factors were diet of the unstressed focal larva (high vs. low quality), food presence/absence (pre- vs. post-food introduction, treated as a repeated measure) and the interaction between diet and food presence/absence. For stressed salamanders, the factors were diet of the stressed larva’s opponent (high vs. low quality), food presence/absence (pre- vs. post-food introduction, treated as a repeated measure) and the interaction between opponent’s diet and food presence/absence. Statistics were calculated with MiniTab v.17.1.0.
Results
Larvae fed the high-quality diet were about 5% larger in body length than larvae fed the low-quality diet (Two-sample t: , , , ). According to the analysis of the residuals from the regression of larval length versus volume, “high-condition” larvae were confirmed to be in significantly better condition than “low-condition larvae” (, , , ) (fig. 1). Body conditions of all larvae were within the range of that observed in natural ponds (Mathis, unpublished data).
For unstressed larvae, high-condition larvae bit their opponents more often than low-condition larvae (ANOVA: , ), with the strongest difference before food was added. Overall, the number of bites increased significantly after food was introduced (, ). There was no significant interaction between diet quality and food presence/absence on number of bites (, ) (fig. 2).
Mean (± SE) body condition residuals between larval salamanders on high-quality diets () and those on low-quality diets ().
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) body condition residuals between larval salamanders on high-quality diets () and those on low-quality diets ().
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) body condition residuals between larval salamanders on high-quality diets () and those on low-quality diets ().
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) number of aggressive bites made by unstressed larvae on high-quality and low-quality diets to their opponents (all on high-quality diets) in the presence or absence of food. Statistics were calculated using the aligned-rank transformation and analysed using a two-way ANOVA with repeated measures.
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) number of aggressive bites made by unstressed larvae on high-quality and low-quality diets to their opponents (all on high-quality diets) in the presence or absence of food. Statistics were calculated using the aligned-rank transformation and analysed using a two-way ANOVA with repeated measures.
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) number of aggressive bites made by unstressed larvae on high-quality and low-quality diets to their opponents (all on high-quality diets) in the presence or absence of food. Statistics were calculated using the aligned-rank transformation and analysed using a two-way ANOVA with repeated measures.
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
For stressed larvae, which all were on high-quality diets, the main effect of opponent’s diet quality was marginally nonsignificant (, ), with a tendency for a higher level of biting when the opponent was of low quality. The level of biting was significantly higher after food was introduced (, ). This difference was primarily due to a significant interaction between the opponent’s diet quality and presence of food, with a high level of biting occurring only when food was present and the opponent was of low-quality (, ) (fig. 3).
Mean (± SE) number of aggressive bites made by stressed larvae (all on high-quality diets) to opponents on high-quality or low-quality diets in the presence or absence of food. Statistics were calculated using the aligned-rank transformation and analysed using a two-way ANOVA with repeated measures.
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) number of aggressive bites made by stressed larvae (all on high-quality diets) to opponents on high-quality or low-quality diets in the presence or absence of food. Statistics were calculated using the aligned-rank transformation and analysed using a two-way ANOVA with repeated measures.
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Mean (± SE) number of aggressive bites made by stressed larvae (all on high-quality diets) to opponents on high-quality or low-quality diets in the presence or absence of food. Statistics were calculated using the aligned-rank transformation and analysed using a two-way ANOVA with repeated measures.
Citation: Amphibia-Reptilia 38, 1 (2017) ; 10.1163/15685381-00003089
Discussion
Interference behaviour is a form of competition that is common in larval ambystomatid salamanders (e.g., Wildy et al., 2001; Mott and Maret, 2011). Intraspecific competition has both short-term and long-term effects on growth rates, body condition, time to metamorphosis and reproduction (Smith, 1990; Scott and Fore, 1995; Doyle, Nolan and Whiteman, 2010). In our study, the factors affecting the performance of interference behaviour included diet quality and stress level of the focal larva, the diet quality of the larva’s opponent, and whether food was present. The effects on levels of biting were complex, with each factor influencing expression of other factors in at least some conditions.
For unstressed larvae, diet quality, which affected both size and condition, was an important factor regardless of whether food was present, with high-condition larvae having significantly higher levels of biting than low-condition larvae when confronted by a high-condition, stressed opponent. Two hypotheses could account for the observed differences in biting frequencies. First, there could be an absolute effect of size/condition, with high-condition larvae biting more frequently than low-condition larvae regardless of the condition of their opponents (e.g., Mathis and Britzke, 1999; Poulos and McCormick, 2015). Second, larvae may bite more in contests with same-condition opponents (i.e., “symmetrical contests”: Maynard Smith, 1982) than in contests with different-condition opponents (“asymmetrical contests”) (Maynard Smith and Parker, 1976). Brunkow and Collins (1998) reported that there were higher levels of aggression in groups of A. tigrinum that were similar in body size. In our study, since the opponents of all unstressed larvae were of high condition, the high-condition larvae were in symmetrical contests and the low-condition larvae were in asymmetrical contests. However, for our study, the data are not consistent with the hypothesis that size/condition relative to opponent is the primary explanation for differences in levels of biting, at least for our range of body size differences. In contrast, relative body size was reported to influence levels of aggression in larval A. talpoideum (Walls and Semlitsch, 1991; Doyle, Nolan and Whiteman, 2010), A. maculatum (Walls and Semlitsch, 1991) and Salamandra salamandra (Reques and Tejedo, 1996), and large larvae of Ambystoma opacum reduced the growth of smaller larvae as a result of interference competition (Smith, 1990). These hypotheses are not mutually exclusive; both absolute size and symmetry/asymmetry of contestants influenced the level of aggression in a terrestrial salamander, Plethodon angusticlavius (Mathis and Britzke, 1999).
The advantage of a high-quality diet appeared to be negated by stress. When food was absent, high-condition stressed larvae had a level of biting opponents that was similar to that of low-condition, unstressed larvae. The effect of stress could be due to handling (e.g., Bliley and Woodley, 2012), to the introduction of the larvae into an unfamiliar testing container (e.g., Langkilde and Shine, 2006), or a combination of both. Although the relative contributions of different stressors for larval salamanders is not known, in scincid lizards, Eulamprus heatwolei, plasma cortisone levels for handling stress were similar to that of exposure to the scent of a predator and less than that for introduction to an unfamiliar chamber (Langkilde and Shine, 2006).
For stressed larvae, half of the opponents were low condition and half were high condition. The effect of opponent condition on the level of biting by stressed larvae depended on whether food was present. In the absence of food, stressed larvae (all high condition) showed a low frequency of biting that was not affected by the diet quality of the opponent. The effect of stress appears equivalent to the effect of low-quality diets, with similar levels of biting by stressed and low-condition larvae. However, when food was present, diet quality relative to the opponent appeared to override the effect of stress, with stressed salamanders biting low-condition opponents more than high-condition opponents, at a level that was only slightly lower than the level of biting by high-condition larvae. The data for the influence of prey presence on interference behaviour by ambystomatid salamanders has not been consistent, including no effect of prey presence on aggression (A. tigrinum and A. maculatum: Mott and Sparling, 2010), higher aggression when food was present (A. tigrinum: Johnson, Bierzychudek and Whiteman, 2003), and lower aggression when food was present (A. opacum: Mott and Sparling, 2010). This variability suggests that the effect of prey presence frequently depends on other interacting variables. For example, individuals in our study were fasted two days before testing and were presumably hungry, so they may have been highly motivated to compete for prey. Costs and benefits of aggression may be different for satiated individuals.
It is unlikely that bites by larvae in our study could have led to cannibalism because size differences between larvae on different diets were relatively small (about 5% of total length). In a field study (Nyman, Wilkinson and Hutcherson, 1993), A. annulatum cannibals were a minimum of 67% larger than their prey. Qualitatively, the bites by larvae in our study were followed by rapid release rather than the bite-and-hold that would be expected for attempts at cannibalism. In addition, bites were sometimes made by low-condition individuals to their high-condition opponents, which is not consistent with the prediction of cannibalism.
In summary, interference behaviour (biting) in larval A. annulatum is influenced by a variety of extrinsic and intrinsic factors that interact in a complex fashion. High-quality diets, which led to larvae that were larger and in better condition, were more aggressive, but acute stress appeared to cancel out this effect, at least in the short term. Overall, larvae were more aggressive when food was present, but the amount of aggression was influenced by condition differences between opponents. The complex balancing of costs and benefits of multiple intrinsic and extrinsic variables in determining levels of interference behaviour is likely common in larval salamanders. Mott and Sparling (2010) reported that interactions between prey abundance, predation risk, and season affected levels of aggression in three other species of larval Ambystoma (A. tigrinum, A. opacum, A. maculatum), and also found difference among species.
Acknowledgements
The experimental protocols were approved by the Missouri State University Institutional Animal Care and Use Committee, and salamanders were collected via a permit from the Missouri Department of Conservation. Emilee Helton assisted with collection and laboratory maintenance. We thank Ben Dalton for his assistance with data analysis and for reviewing the manuscript along with two anonymous referees.
References
Bliley J.M., Woodley S.K. (2012): The effects of repeated handling and corticosterone treatment on behavior in an amphibian (Ocoee salamander: Desmognathus ocoee). Physiol. Behav. 105: 1132-1139.
Brunkow P.E., Collins J.P. (1998): Group size structure affects patterns of aggression in larval salamanders. Behav. Ecol. 9: 508-514.
Cupp P.V. Jr. (1980): Territoriality in the green salamander, Aneides aeneus. Copeia 1980: 463-468.
Doyle J.M., Nolan J.R., Whiteman H.H. (2010): Effects of relative size on growth rate and time to metamorphosis in mole salamanders (Ambystoma talpoideum). J. Herpetol. 44: 601-609.
Echaubard P., Little K., Pauli B., Lesbarrères D. (2010): Context-dependent effects of ranaviral infection on northern leopard frog life history traits. PLoS ONE 5: e13723. DOI:10.1371/journal.pone.0013723.
Fox L.R. (1975): Cannibalism in natural populations. Annu. Rev. Ecol. Syst. 6: 87-106.
Gammie S.C., Stevenson S.A. (2006): Effects of daily and acute restraint stress during lactation on maternal aggression and behavior in mice. Stress 9: 171-180.
Higgins J.J., Tashtoush S. (1994): An aligned rank transform test for interaction. Nonlinear World 1: 201-211.
Jefferson D.M., Ferrari M., Mathis A., Hobson K., Britzke E., Blaustein A., Crane A., Chivers D. (2014): Shifty salamanders: transient trophic polymorphism and cannibalism within natural populations of larval ambystomatid salamanders. Front. Zool. 1: 76.
Johnson E.B., Bierzychudek P., Whiteman H.H. (2003): Potential of prey size and type to affect foraging asymmetries in tiger salamander (Ambystoma tigrinum nebulosum) larvae. Can. J. Zool. 81: 1726-1735.
Johnson T.R. (2000): Amphibians and Reptiles of Missouri, 2nd Edition. Missouri Department of Conservation, Jefferson City.
Kishida O., Tezuka A., Ikeda A., Takatsu K., Michimae H. (2015): Adaptive acceleration in growth and development of salamander hatchlings in cannibalistic situations. Funct. Ecol. 29: 469-478.
Langkilde T., Shine R. (2006): How much stress do researchers inflict on their study animals? A case study using a scincid lizard, Eulamprus heatwolei. J. Exp. Biol. 209: 1035-1043.
Magellan K., Kaiser H. (2010): The function of aggression in the swordtail, Xiphophorus helleri: resource defence. J. Ethol. 28: 239-244.
Martel G., Dill L.M. (1993): Feeding and aggressive behaviours in juvenile coho salmon (Oncorhynchus kisutch) under chemically-mediated risk of predation. Behav. Ecol. Sociobiol. 32: 365-370.
Mathis A., Britzke E. (1999): Asymmetries in body size and experience influence agonistic displays in the zigzag salamander, Plethodon angusticlavius. Herpetologica 55: 344-352.
Mathis A., Jaeger R.G., Keen W.H., Ducey P.K., Walls S.C., Buchanan B.W. (1995): Aggression and territoriality by salamanders and a comparison with the territorial behavior of frogs. In: Amphibian Biology, vol. 2, p. 633-676. Heatwole H., Sullivan B.K., Eds, Surrey Beatty & Sons, Chipping Norton, Australia.
Mathis A., Schmidt D.W., Medley K.A. (2000): The influence of residency status on agonistic behavior of male and female Ozark zigzag salamanders Plethodon angusticlavius. Am. Mid. Nat. 143: 245-249.
Maynard Smith J., Parker G.A. (1976): The logic of asymmetric contests. Anim. Behav. 24: 159-175.
Maynard Smith J. (1982): Evolution and the Theory of Games. Cambridge University Press, Cambridge.
Mineur Y.S., Prasol D.J., Belzung C., Crusio W.E. (2003): Agonistic behavior and unpredictable chronic mild stress in mice. Behav. Genet. 33: 513-519.
Mott C.L., Maret T.J. (2011): Species-specific patterns of agonistic behavior among larvae of three syntopic species of ambystomatid salamanders. Copeia 2011: 9-17.
Mott C.L., Sparling D.W. (2010): Seasonal trends in aggression among sympatric larval salamanders: the roles of habitat-mediation and behavioural conservatism. Behaviour 147: 1327-1353.
Nyman S., Wilkinson R.F., Hutcherson J.E. (1993): Cannibalism and size relations in a cohort of larval ringed salamanders (Ambystoma annulatum). J. Herpetol. 27: 78-84.
Peterson C.L., Wilkinson R.F., Moll D., Holder T. (1991): Prematamorphic survival of Ambystoma annulatum. Herpetologica 47: 96-100.
Petranka J.W. (1989): Density-dependent growth and survival of larval Ambystoma: evidence from whole-pond manipulations. Ecology 70: 1752-1767.
Poulos D.E., McCormick M.I. (2015): Asymmetries in body condition and order of arrival influence competitive ability and survival in a coral reef fish. Oecologia 179: 719-728.
Prenter J., Taylor P.W., Elwood R.W. (2008): Larger body size for winning and larger swords for winning quickly in swordtail males, Xiphophorus helleri. Anim. Behav. 75: 1981-1987.
Reques R., Tejedo M. (1996): Intraspecific aggressive behaviour in fire salamander larvae (Salamandra salamandra): the effects of density and body size. Herpetol. J. 6: 15-19.
Ricciardella L.F., Bliley J.M., Feth C.C., Woodley S.K. (2010): Acute stressors increase plasma corticosterone and decrease locomotor activity in a terrestrial salamander (Desmognathus ochrophaeus). Physiol. Behav. 101: 81-86.
Richards A.G. (1951): The Integument of Arthropods: the Chemical Compounds and Their Properties, the Anatomy and Development, and the Permeability. Univ. Minnesota Press, Minneapolis.
Sacchi R., Pupin F., Gentilli A., Rubolini D., Scali S., Fasola M., Galeotti P. (2009): Male-male combats in a polymorphic lizard: residency and size, but not color, affect fighting rules and contest outcome. Aggress. Behav. 35: 274-283.
Scott D.E., Fore M.R. (1995): The effect of food limitation on lipid levels, growth, and reproduction in the marbled salamander, Ambystoma opacum. Herpetologica 51: 462-471.
Simon M.P., Toft C.A. (1991): Diet specialization in small vertebrates: mite-eating in frogs. Oikos 61: 263-278.
Smith C.K. (1990): Effects of variation in body size on intraspecific competition among larval salamanders. Ecology 71: 1777-1788.
Sogard S.M. (1997): Size-selective mortality in the juvenile stage of teleost fishes: a review. Bull. Mar. Sci. 60: 1129-1157.
Stenhouse S.L., Hairston N.G., Cobey A.E. (1983): Predation and competition in Ambystoma larvae: field and laboratory experiments. J. Herpetol. 17: 210-220.
Urban M.C. (2008): Salamander evolution across a latitudinal cline in gape-limited predation risk. Oikos 117: 1037-1049.
Wack C.L., Ratay M.K., Woodley S.K. (2013): Effects of corticosterone on locomotory activity in red-legged salamanders. Herpetologica 69: 118-126.
Walls S.C., Jaeger R.G. (1987): Aggression and exploitation as mechanisms of competition in larval salamanders. Can. J. Zool. 65: 2938-2944.
Walls S.C., Semlitsch R.G. (1991): Visual and movement displays function as agonistic behavior in larval salamanders. Copeia 1991: 936-942.
Wildy E.L., Chivers D.P., Kiesecker J.M., Blaustein A.R. (2001): The effects of food level and conspecific density on biting and cannibalism in larval long-toed salamanders, Ambystoma macrodactylum. Oecologia 128: 202-209.
Woodley S.K., Lacy E.L. (2010): An acute stressor alters steroid hormone levels and activity but not sexual behavior in male and female Ocoee salamanders (Desmognathus ocoee). Horm. Behav. 58: 427-432.
Woodley S.K., Porter B.A. (2016): Handling stress increases expression of male sexual behaviour in an amphibian with an explosive mating strategy. J. Zool. 298: 178-182.
Footnotes
Associate Editor: Blake Hossack.