Anti-predator behavior in amphibian larvae

[Alyrium Denryle]

Figured I would post an old term paper for public enjoyment and discussion. That and I am bored as all hell, and wanted to see if anyone on earth is interested in amphibian behavior.



Costs and Life History Tradeoffs of Antipredator Strategies in Amphibians
By Benjamin Allen







Introduction

The predation risk at any given time is highest within the attack range of a predator, and is equal to the probability of a successful attack by said predator. If this is examined from the perspective of optimization, one would expect that a prey species will react primarily to this momentary predation risk. This is because by only responding when threats are imminent, they do not incur additional costs of defense, while still responding maximally to the predator (Fraker 2008). However, most of the time, the prey species is lacking the information necessary to completely optimize their strategy (Sih, Ziemba & Harding 2000). Because information is limited, the prey species is left with a choice. They can either incur a higher cost to defend themselves from predators, or they can incur the predation risk without mitigation. These are of course two extreme cases, but this sets up trade-offs between predation mitigation and the costs of anti-predator defense. Aquatic-breeding amphibians are interesting this respect. They possess complex life cycles that include ontogenic shifts in every aspect of their ecology from their habitat to position on the trophic web. They experience different predators and are under different selective pressures during their larval period than they do as adults. There are several different types of anti-predator strategies that amphibians engage in. Morphological defenses consist of things like increased tail depth for increased swimming performance (Lardner 2000) physiological defenses consist of secreted toxins (Brodie et al 1978), and behavioral defenses consist of tactics such as predator avoidance and alarm calls (Shelly and Werner 1990). There are also shifts in life history to minimize exposure to the highest predation risk. The trade-offs involved in all of these can lead to secondary effects on life history such as reduction in metamorph size and performance (Lardner 2000) and increases in larval period (Skelly 1992). It will be the anti-predator strategies of larval amphibians and subsequent life history effects that will the primary subject of this review. It will be done in a few stages: the first step will be to examine how amphibians assess predation risk; the second will be to evaluate modes of response to predation with a mind to what the costs of each mode could be; lastly an explicit look at the costs of predator-mitigation and the effects on life history traits.

Methods

In order to evaluate these questions, Web of Science was searched using the keywords “Tadpole+ Predator”, “Anuran+Predator”, and “Amphibian+Predator” and sorted by relevance. Articles were selected, and their works cited pages searched for additional references.

Results and Discussion

Primarily amphibian larvae use chemical cues from predatory organisms in order to assess predation risk (Keisecker et al 1996). However these signals are not always reliable. Predators move, and chemical signals diffuse through the water column and degrade (Peacor 2006). There are several predictions that one can deduce from this. One would expect that after exposure, larval amphibians would return to their previous behavior in order to mitigate the costs of anti-predator defense, but would do so with some lag-time. This hypothesis was evaluated by Fraker (2008) using Lithobates clamitans and its response to Anax predators. He found that the perceived risk (as measured by lag-time to control activity levels after cessation of cue exposure) to predators varied with the concentration of the cue, cue exposure length, and was inversely related to tadpole size. This indicates that perceived risk, and time since last exposure mediates anti-predator behavior, and helps to obviate some of the costs involved.

There are other ways in which an amphibian larva might reduce risk. For example: by assessing predator density via chemical cues, and via the presence of conspecific alarm cues. Anholt et al.(1996) evaluated this by subjecting Bufo americanus tadpoles to a series of experiments that measured tadpole response to conspecific alarm cues (cues given off by depredated tadpoles), cues from high predator density, tadpole size, and food levels. They found that tadpoles responded by physically avoiding predators. They also found that tadpoles reduced activity with higher predator density, and higher food levels. Additionally tadpoles which were smaller had a stronger reaction to predators. Reactions to conspecific alarm cues were non-significant in this case. These data indicate that tadpoles are assessing predation risk in terms of predator density, and their own size, with higher predator density, and smaller tadpole size leading to an increased reaction as opposed to lower density or larger tadpoles. The reason for this is of course that larger tadpoles have a higher probability of being able to escape a predator, and higher predator density indicates a higher attack probability in any given area.

It is also worth noting that responses are risk specific. Keisecker et al (1996) exposed Bufo boreas tadpoles to visual and chemical cues from three predators: Notonecta, Lethocerus americanus, and Thamnophis sirtalis, as well as two potential predators; Tarichia granulose, and the trout, Oncorhynchus mykiss, These last two find toad tadpoles unpalatable in the wild due to skin toxins produced by the toads. The toads responded by reducing activity and physically avoiding predators which found them palatable, but showed no response to the two potential predators. This was done in response to chemical cues, and not visual stimulus. This indicates that the tadpoles can discriminate between predators which are of high risk, and low risk.

So how do they typically respond to predators? As has been mentioned before, amphibian larvae use a combination of behavioral, morphological and chemical defenses. Additionally, as has just been demonstrated in Keisecker et al, one defense can obviate the need for others. In this case, chemical defenses can reduce the need for behavioral defenses against predators which the tadpole cannot easily out-maneuver behaviorally. Additionally, defenses can be fixed (as is the case in many chemical defenses), they can be plastic, or heritable and inducible, as is the case in many morphological or behavioral defenses (Smith and Buskirk 1995). Plastic defenses such as changes in tail depth and alterations in behavior are expected to be more prevalent in species with either a wide habitat preference, or a single preferred habitat which itself has a highly variable predator composition (Lardner 2000). For example, semi-permanent ponds can accumulate a large number of invertebrate and amphibian predators that are wiped out when the pond eventually dries, creating a wide range of temporal variation in the predation risk in such ponds. Amphibians which breed in these ponds would typically possess phenotypically plastic defenses against predators so that they can avoid predators when present, and not pay the cost of defense when the predators are not present. Amphibians in vernal pools generally do not defend themselves significantly because of the risk of pond drying outweighs the risk of predation, and thus the costs of defense are not justified. Amphibians in permanent water, because predation risk is constant, tend to possess fixed or otherwise heritable defenses which are optimized to local predation risk (Lardner 2000, Smith and Van Buskirk 1995).

Lardner (2000) tested this hypothesis with seven species of European anurans, and found that these predictions were largely true. The anurans which bred in spatiotemporally variable habitats (Rana temporaria,Rana arvalis, Rana dalmatina and Hyla arborea) responded by reducing growth rates, and increasing tail depth, which increased swimming performance. The permanent pond breeders tended to not have as plastic responses, and only Bufo bufo reduced its growth rate, Paleobates fuscus did not exhibit any predator mediated pasticity. Neither did the ephemeral breeding Bufo calamita.

So why not always exhibit anti-predator defenses? Anti-predator defenses incur costs (Skelly 1992, Van Buskirk 2001, Lardner 2000). This cost of often manifested in life history traits such as an increase in larval period (Skelly 1992, Van Buskirk 2001, Van Buskirk and Schmidt 2001), or reduced metamorph size (Lardner 2000. Relyea 2001), both of which can lead to increased mortality, either from desiccation or predation risk from predators that feed on metamorphs (Govindarajulu 2005). This has been alluded to above. Many amphibians breed in temporary water bodies (Storfer 1999, lardner 2000, Relyea 2001), and some, such as new world spadefoots, have such a limited time to metamorphose before pond dying, that any increase in larval period would be selected against, regardless of the mortality risk from predation. This is because while a predator can remove an individual from a tadpole population, desiccation can remove and entire clutch of eggs, and create a little tadpole pancake. As a result, selection is essentially wagering individual tadpole mortality rate against mortality for the entire clutch of eggs. The question now becomes, what part of an amphibian’s life cycle is most vulnerable to predation? In other words, where does mortality have the largest effect on reproductive success? This was evaluated by Govindarajulu (2005) who found that in bullfrogs, the largest influences on population size were early metamorph survival, and the proportion of tadpoles that metamorphosed early (IE. They did not overwinter in ponds). Now, because many amphibian predators are gape limited larger metamporphs are more likely to survive predation attempts, and be subjected to fewer in the first place. As a result, it would be expected that reductions in metamorph size and performance would be compensated for during development, probably through an increase in larval period. This was found by Lardner (2000) in most species surveyed which exhibited plasticity (all but Rana temporaria). This was also found by Capellan and Nicieza (2007). In other words, the larvae respond to predators by decreasing activity (and thus feeding rate) and potentially developing morphological defenses against predation. They trade this for a longer larval period, which allows them to avoid harming their chances of survival as metamorphs, in exchange for an increased risk of desiccation.

There are three things which need to be addressed. Through most of this review, heritable defenses (as opposed to plastic ones) have been overlooked, as many of the papers reviewed implicitly or explicitly assumed that the defenses involved are phenotypically plastic. It is entirely possible that a defensive response, or anti-predator adaptation can be highly heritable (IE. Different populations will respond differently in a common garden) but still be induced. It has also been the case that gene flow has been overlooked, and so has the possibility for non-adaptive phenotypes. Gene flow has the potential to create sub-optimal adaptation. Phenotypic plasticity will only be maintained if there is selection on it. So if a species is fairly philopatric, and the individuals present in particular populations do not experience variability in predation regimes, but different populations experience differing levels of predation, then what would be expected is that individual populations will be under positive selection to adapt specifically to the predation regime which they encounter. Gene flow would be expected to interfere with this, and proximity to populations which experience a different predation regime would be expected to have a negative impact on local adaptation. This was found by Storfer and Sih (1998) and Storfer et al (1999). They found that larvae of the stream breeding salamander Ambystoma barbouri exhibit two types of defenses: an induced behavioral defense (activity and feeding reduction) and cryptic coloration. Both of these responses are heritable, and depend on the population in question. Salamanders which breed in ephemeral streams have a limited time to reach metamorphosis, and thus are active and have darker coloration (presumably for thermoregulatory purposes), they also do not experience predation by fish. Salamanders which live in permanent streams experience fish, and are lighter in color (to better match the background color) and reduce activity. These phenotypes however depended on the fish-depredated population’s proximity to fish-free populations, and did not become the same response in a common garden (or rather, common aquarium) experiment. The closer they (the depredated population) were to a fish free population, the less effective the behavioral response to predation, and the darker the color.

There are a few criticisms possible with this logic. It could be argued that a population under predation at a given life history stage would want to get out of that period as fast as possible, and as a result would do something to decrease their metamorphosis time. This is seen in garter snakes (Shine et al 2001). Ecomorphs under strong predation grow and mature faster than a geographically close ecomorph under weaker predation pressure (Bronikowski 1999). However this is what happens when theory does not take into account the different types of predation occur. Organisms that prey upon tadpoles are much different than those that prey upon snakes. Many tadpole predators are either not gape-limited, or are simply too large for the predator to ever become large enough to be immune to predation from (pers observation). Many of them also feed on metamorphs (such as the aforementioned garter snakes). Unlike the snakes which can grow large enough and fast enough to literally outgrow many of their predators (garter paper). The best strategy is to develop defenses against predators which do not rely on size insofar as size is not related to performance. Particularly because the increased activity associated with increasing growth rate will also increase the risk of predation. These defenses incur costs in terms of growth rate and time to metamorphosis.

Conclusion
In conclusion, the responses of amphibian larvae to predators tend to be optimized toward the specific threats that they face in order to minimize life history costs. However, gene flow can negatively impact this optimization for non-plastic defenses. More work needs to be done in order to fully ferret out how environmental variability and habitat subdivision, as well as novel predators play into the evolution of anti-predator strategies.














References
Anholt, BR., Skelly, DK., Werner, EE. Factors Modifying Antipredator Behavior in Larval Toads. Herpetologica, 52:301-313.

Anholt, BR., Werner, EE., Skelly, DK.2000.Effect of Food and Predators on the Activity of Four Larval Ranid Frogs. Ecology, 81:3509-3521.

Brodie, ED., Formanowicz, DR., Brodie, ED.1978. Development of Noxiousness of Bufo americanus tadpoles to aquatic insect predators. Herpetologica, 34:302-306.

Bronikiowski, A., Arnold, S. 1999. The Evolutionary Ecology of Life History Variation in the Garter Snake Thamnophis elegans. Evolution, 54:1760-1767.
Capellan, E., Nicieza, A. 2007. Trade-offs across life stages: does predator-induced hatching plasticity reduce anuran post-metamorphic performance?. Evolutionary Ecology, 21:445-458.

Fraker, ME. 2008. The dynamics of predation risk assessment: responses of anuran larvae to chemical cues of predators. JOURNAL OF ANIMAL ECOLOGY, 77: 638-645.

Govindarajulu, P., Altwegg, R., Anholt, BR.2005.Matrix Model Investigation of Invasive Species Control: Bullfrogs in Vancouver Island. Ecological Applications, 15:2161- 2170.

Kats, LB., Petranka, JW., Sih, A.1988.Antipredator Defenses and the Persistence of Amphibian larvae with Fishes. Ecology, 69:1865-1870.
Keisecker, JM., Chivers, DP., Blaustein, AR. 1996. The Use of Chemical

Cues in Predator Recognition by Western Toad Tadpoles. Animal behavior, 52:1237-1245.

Larnder, B. 2000. Morphological and Life History Responses to Predators in Larvae of Seven Anurans. Oikos, 88:169-180.

Peacor, S. 2006. Behavioral Response of Bullfrog Tadpoles to Chemical Cues of Predation Risk are Affected by Cue Age and Water Source. Hydrobiologica, 573:39-44.

Relyea, R. 2001. Morphological and behavioral plasticity of larval anurans in response to different predators. ECOLOGY, 82:523-540.

Relyea, R.2001 The relationship between predation risk and antipredator responses in larval anurans. ECOLOGY, 82:541-554.

Relyea, R. 2001. The lasting effects of adaptive plasticity: Predator-induced tadpoles become long-legged frogs. ECOLOGY, 82:1947-1955.

Shine, R.,Lemaster, M., More, I., Olsson, M., Mason,R. 2001. Bumpus in the Snake Den:Effects of Sex, Size, and Body Condition on Mortality of Red-Sided Garter Snakes (Thamnophis sirtalis parietalis). Evolution, 55:598-604

Sih, A., Ziemba, R., Harding, KC. 2000. New Insights on How Temporal Variation in Predation Risk Shapes Prey Behavior. Trends in Ecology and Evolution, 15:3-4.

Skelly, D.1992. Field Evidencefor a Cost of Behavioral Antipredator Response in a Larval Amphibian. Ecology, 73:704-708.

Smith, DC., Van Buskirk, J. 1995.Phenotypic Design, Plasticity, And Ecological Performance In 2 Tadpole Species. American Naturalist, 145:211-233.

Storfer, A., Cross, Jonra., Rush, V., Caruso, J. 1999. Adaptive Coloration and Gene Flow as a Constraint to Local Adaptation in the Streamside Salamander, Ambystoma barbouri. Evolution, 53:889-898.

Storfer, A., Sih, A. 1998. Gene Flow and Ineffective Antipredator Behavior in a Stream Breeding Salamander. Evolution 52:558-565.

Van Buskirk, J., McCollum, S. 2000. Functional mechanisms of an inducible defence in tadpoles: morphology and behaviour influence mortality risk from predation. Journal of Evolutionary Biology, 13:336-347.

Van Buskirk, J., McCollum, S. 2000. Influence of tail shape on tadpole swimming performance. Journal of Experimental Biology, 203:2149-2158.

Van Buskirk, J. 2000. The costs of an inducible defense in anuran larvae. ECOLOGY, 81:2813-2821.

Van Buskirk, J., Schmidt, B. 2000. Predator-induced phenotypic plasticity in larval newts: Trade-offs, selection, and variation in nature. Ecology, 81:3009-3028.