Printer Friendly

Space Use of Predatory Larval Dragonflies and Tadpole Prey in Response to Chemical Cues of Predation.


How individuals make movement decisions and distribute themselves spatially determines their rates of direct interactions with other species (Sih, 2005; Hammond et al., 2007; Laundre, 2010; Fraker and Luttbeg, 2012), as well as the structure of various indirect interactions (Werner and Peacor, 2003). In general mobile prey should minimize the probability of predation mortality while maximizing resource intake, and predators should maximize their prey capture rate (Gilliam and Fraser, 1987). However, individuals rarely have complete information on the condition of their environment (Lima and Steury, 2005; Luttbeg and Trussell, 2013). Therefore, identifying what information individuals use to assess costs and benefits and then how they respond to it spatially is necessary to understand how predator-prey interactions operate.

Chemical cues are a key source of information, particularly in aquatic systems (Ferrari et al., 2010). Chemicals are released by predators and prey before, during, and after predation events (Schoeppner and Relyea, 2009; Ferrari et al., 2010; Hettyey et al., 2015; Shaffery and Relyea, 2016). Prey may detect predators through chemicals inadvertently released by the predators (e.g., kairomones), chemicals released by other prey either during or after attack or capture (e.g., alarm cues and damage-released cues), and chemicals released by predators after digesting prey (e.g., metabolites). Likewise, predators may be able to detect these chemicals and use them as information sources.

Several studies have shown aquatic prey spatially avoid chemical cues produced during predation events (e.g., Relyea and Werner, 1999; Turner and Montgomery, 2003; Friesen and Chivers, 2006; Wisenden, 2008) and aquatic predators may use predation event cues to forage (e.g., Lonnstedt and McCormick, 2015). However, how individuals respond spatially to different types of cues remains unclear. Individuals are likely to encounter different types of chemical cues singly and in combination as they move through their habitat. Whether movement decisions are adaptive depends on how individuals assess and respond to different cue types. Therefore, it seems likely different cues have different informational values. For example kairomones alone indicate the presence of a predator that is not currently feeding but may pose a risk of attack if encountered. Alarm- or damage-released cues indicate a recent attack or successful prey capture but provide no information about the species of predator. Predator-released metabolites indicate the presence of a predator and may provide information on the prey species. Depending on the predator species, metabolites may also indicate lower risk from that particular individual if successful prey capture reduces a predator's subsequent motivation. A further complication is that individuals of many species can quickly learn to associate particular chemicals with risk (or another environmental state) through associative learning or conditioning (Chivers and Smith, 1998).

In this study we assessed how predators and prey responded spatially with exposure to various chemical cues of predation using a system of predatory common green darner dragonfly (Anaxjunius) larvae and predator-naive green frog (Lithobates clamitans) tadpoles. Given previous studies found antipredator behavior decreases with size in tadpoles (e.g., Fraker, 2008), we tested two size classes of tadpoles. Because we hypothesized Anax feeding state may affect information use, we also tested both recently fed and starved Anax. Both species were exposed to dragonfly kairomones, tadpole alarm cues, or the full set of predation event cues. While previous studies using these species have found strong behavioral (e.g., activity level, refuge use) and other phenotypic (e.g., morphological) responses in tadpoles to chemical cue exposure (e.g., Relyea, 2001; Fraker, 2008), they did not record space use in response to cue type or measure dragonfly behavior. Improving our mechanistic understanding of predator and prey information use and decision making will help provide a foundation for future studies exploring the spatial dynamics of predator-prey systems.



Freshly-laid green frog egg masses and late-instar Anax larvae were collected during May (green frog egg masses) and June (dragonflies) 2016 from restored wetlands in which they co- occur on the property of the Ohio Department of Natural Resources-Division of Wildlife's State Fish Hatchery located in Hebron, Ohio (39[degrees]56'25"N, 82[degrees]30'54"W). Specimens were transported to The Ohio State University's Aquatic Ecology Laboratory for housing and use in experiments (approved under OSU IACUC protocol # 2016A00000028). Egg masses were cultured outdoors in fiberglass screen-covered 75 L wading pools filled with aged dechlorinated tap water and inoculated with plankton from native ponds. Once hatched tadpoles were fed rabbit chow (Purina, St. Louis, Missouri) ad libitum. The tadpoles were raised in predator-free pools (i.e., predator-naive) to remove confounding effects of prior predation and learning on future behavior. Anax were housed in plastic containers filled with 400 mL of aged, dechlorinated tap water; water was changed weekly to maintain water quality. Anax were fed approximately 100 mg of live tadpoles three times per week and held without food for at least 48 h prior to producing chemical cues for experiments. Anax scheduled for use in an experiment always had their water changed 48 h prior to the experiment.

We conducted our experiments indoors in 75 L mesocosm pools (~1 m diameter) filled with aged dechlorinated tap water to a depth of 20 cm. The pools were located on the floor of a well-ventilated room with fluorescent full spectrum lights set to a 14:10 light:dark schedule and a constant temperature of 21 C. We sectioned the experimental pools into four quadrants of equal size (Fig. 1). Each quadrant contained one algal wafer disc (i.e., resource patch) and one completely submerged opaque plastic cup covered with a fiberglass window screen (i.e., predator cage in predator treatments or empty cup in predator-free treatments). The mouth of the cup was facing the center of the pool with the back of the cup next to the pool wall. The algal wafer was placed in front of the cup mouth. One quadrant was randomly designated as the treatment quadrant. It received the caged dragonfly, chemical cue, or control and was the focal quadrant for our behavioral observations (described below). Each pool also received strips of fiberglass screen within each quadrant to serve as structure.

We observed and recorded the number and distribution of tadpoles and dragonflies among quadrants within the pools 30 min before the start of each experimental trial to ensure a random initial distribution. We then observed and recorded distributions once every 30 min during a 2 h period after treatment application (n = four observations). We only used individual tadpoles in one replicate, whereas individual dragonflies were used multiple times (with at least 72 h between trials).

Experiment 1: prey (tadpole) space use.--The prey experiment measured the spatial response of two size classes of tadpoles exposed to five treatments of chemical cues. The size classes were: 50 mg (40-60 mg) and 150 mg (100-200 mg). Tadpoles from the 50 mg size class were at Gosner stage 26, while the 150 mg size class were at Gosner stages 26-27 (Gosner, 1960). Both size classes of tadpoles came from the same sets of egg masses; treatments using the small and large size classes were conducted 3 wk apart during July 2016. The five chemical cue treatments were: (1) caged, feeding Anax (dragonfly kairomones, tadpole alarm cues, possibly metabolites); (2) caged, unfed Anax (kairomones only); (3) tadpole-Triton cue (alarm cues only; Fraker et al., 2009); (4) Triton (control for alarm cues); and (5) water only (control). We replicated each tadpole size class and cue treatment combination four to six times. Each size class was run as a temporal block as tadpoles grew into the size class (i.e., two experimental periods).

For each size class, we mixed tadpoles from multiple egg masses (n = 5) prior to randomly sorting into groups of 20, which were then placed into the experimental pools to acclimate overnight. The following day, we simultaneously applied all of the chemical cue treatments during the morning. We applied the two caged Anax treatments by placing one randomly selected Anax into the treatment quadrant's plastic cup. The cup opening was covered by fiberglass window screen so that the Anax could not escape and hunt the tadpoles, but chemical cues could pass through. The Anax that were fed were provided one tadpole corresponding to the same size class as the observed tadpoles (50 mg or 150 mg). Prior to adding the Anax and tadpole into the experimental pool, an observer confirmed the Anax had captured the tadpole. To create the tadpole alarm cue, we followed Fraker et al. (2009), who found ranid tadpole alarm cue (pheromone) is contained in tadpole skin cell vesicles and is only released via an active secretory process. Briefly, we anaesthetized tadpoles corresponding to the same size class as the observed tadpoles (50 mg or 150 mg) in 0.025% benzocaine, then added to a solution of 1% Triton X-100 (Sigma-Aldrich, St. Louis, Missouri, U.S.A) and 100 mL of de-ionized water (one tadpole per replicate and 100 mL water per tadpole). The detergent Triton X-100 solubilizes cell membranes and releases the alarm cue into solution. We sonicated the tadpoles, water, and Triton X-100 mixture for 30 s (Fisher Scientific Homogenizer Power Gen, Model 125, Fisher Scientific, Pittsburgh, Pennsylvania). We then slowly injected 100 mL of the solution into each pool next to the mouth of the cup in the treatment quadrant. In addition to the cue treatments, we created two controls: (1) a 100 mL solution containing only 1% Triton X-100, and (2) 100 mL water.

Experiment 2: predator (dragonfly) space use.--The predator experiment measured the spatial response of individual Anax larvae in two feeding states (recently fed or unfed for 48 h) exposed to six treatments of chemical cues. Individual Awaacwere tested because Anax can be cannibalistic, which may affect their space use and other behaviors (Anderson, 2016). The six chemical cue treatments were: (1) caged, feeding Anax (dragonfly kairomones, tadpole alarm cues, possibly metabolites); (2) caged, unfed Anax (kairomones only); (3) feeding Anax cue without caged Anax (kairomones and alarm cues); (4) unfed Anax cue without caged Anax (kairomones only); (5) tadpole-Triton cue (alarm cues only); and (6) water only (control). Because larval Anax have strong vision, we were concerned caged Anax might alter the response of the focal Anax to chemical cues. To separate the effect of the presence of the Anax from the effect of the chemical cue, we added extra treatments in which the chemical cue was created outside of the experimental pools and then added. We replicated each dragonfly feeding state (fed, unfed) and chemical cue combination six times.

During the morning of each experiment, we placed individual Anax in the center of the experimental pools and allowed to acclimate for at least 1 h. Anax were either unfed (i.e., starved for 48 h prior to use) or fed (i.e., fed one tadpole just prior to addition to a pool). We prepared the chemical cue treatments using the same procedures described above, except for the feeding and starved Anax cues without caged Anax (cues three and four). These were produced 1 h before the experiment by placing Anax into fresh cups of water, then feeding or not feeding them. We then slowly injected the water (chemical cue) into the experimental pools in the treatment quadrant next to the cup mouth.


We calculated space use as the mean proportion of individuals in the treatment quadrant over the observation period for each replicate pool. For the tadpole experiment (Experiment 1), we used the proportion of the 20 individuals in the treatment quadrant at each observation time as the value for each replicate. Because preliminary analyses indicated no temporal patterns during the 2 h experimental period (data not shown), we pooled the four observations of a treatment replicate. For each experiment we separately used two-way analyses of variance to compare that experiment's treatments. Our models looked at the effect of the main factors (chemical cue and tadpole size class or dragonfly feeding state) and their interaction. When the interaction term was not significant, we reran the tests with the interaction term left out. We used Tukey's honestly significant difference tests to identify treatment differences when a main factor or the interaction term was found significant. All data met assumptions for normality using the Kolmogorov-Smirnov test (all P > 0.05) and homogenous variances using Levene's test. Statistical analyses were performed using JMP Pro 12.2 (SAS Institute Inc.).



Preliminary analyses indicated tadpole distributions did not differ from random prior to treatment application. Neither size class (F = 0.14, d.f. = 1, 42, P = 0.71) nor the interaction between size class and chemical cue treatment (F = 0.09, d.f. = 4, 42, P = 0.50) had a significant effect on tadpole space use, whereas chemical cue treatment alone did have a significant effect on space use (F = 6.62, d.f. = 4, 42, P = 0.0003; Fig. 2A). Therefore, in a subsequent analysis, we excluded the interaction term and as well as the effect of size class, and pooled all individuals based on chemical cue treatment. We again found a significant effect of chemical cue treatment on space use (F = 7.49; d.f. = 4, 47, P < 0.0001). Tadpole avoidance was strongest in the caged and fed Anax treatment compared to the water control (P < 0.0001). In that treatment tadpoles generally favored the quadrant opposite the treatment quadrant (i.e., moved to Quadrant 3 away from Quadrant 1; the mean proportions in each quadrant during the observation period pooled over size classes and replicates were: Q1 - 7%, Q2 - 26%, Q3 - 46%, and Q4 - 22%). By contrast the caged and unfed Anax, tadpole-Triton, and Triton control treatments did not differ from the water control. However, inspection of the 95% confidence intervals showed the tadpole-Triton treatment also did not encompass the 0.25 proportion expected from a random distribution, suggesting that tadpoles in this treatment exhibited some avoidance of the tadpole-Triton cue

(i.e. tadpole alarm cue).


Anax space use did not differ from random prior to treatment application. Likewise, feeding state and chemical cue exposure did not have significant main effects or an interaction effect on space use (feeding state: F = 0.006, d.f. = 1, 71, P = 0.94; chemical cue: F = 0.102, d.f. = 5, 71, P = 0.99; feeding state by chemical cue: F = 0.036, d.f. = 5, 71, P = 0.99; Fig. 2B). The main effects remained insignificant after reanalyzing the data with the interaction term removed.


Our results show that predator-naive tadpoles reacted with spatial antipredator behavior only when exposed to both Anax kairomones and conspecific alarm cues together (i.e., a typical predation event; Fig. 2A). Tadpoles did not display spatial antipredator behavior in response to either kairomones or alarm cues alone when compared to the control (although tadpoles did exhibit some avoidance of the alarm cue alone when compared to the expected random distribution). Several possible explanations for these results exist. First, because spatial avoidance (or other antipredator behaviors) may result in lost foraging opportunities, it may be costly to respond to information sources that are not sufficiently reliable indicators of predator location and risk. A combination of predator kairomones and tadpole alarm cues should indicate a recent predation event (i.e., an actively feeding predator nearby), which may be more reliable than cues of a predator or conspecific alone. The weaker avoidance of the tadpole alarm cue is consistent with it providing less information than the combined predator-sourced and prey-sourced chemical cues. Second, we focused on space use in this experiment; however, other antipredator behaviors such as activity reduction may have also been used at the same time (e.g., Fraker, 2008). We focused on space use because it has been less studied than activity level or refuge use and is the directly relevant to the risk of encountering a predator. In future studies recording all aspects of the antipredator response of prey may be necessary to accurately interpret its function and consequences. For example, in one location and at one time, a prey's best response to a perceived increase in predation risk may be either to move away or to become inactive. If the prey stays, it will have to make a temporal assessment and decision of when it is safe enough to resume activity. If the prey moves away, it may be able to remain active, but it also may encounter a different predator. Third, to recognize predator kairomones as indicators of risk, some species require associative learning (e.g., Mirza, et al, 2006; Albecker and Vance-Chalcraft, 2015). If green frog tadpoles do not innately recognize Anax kairomones as indicating risk, then repeated exposures may be necessary before tadpoles would exhibit avoidance behavior. If associative learning is required in green frog tadpoles, we can hypothesize mortality may be high initially (until learning has occurred) given tadpoles would not necessarily avoid Anax that had not fed recently, but still pose a risk. In fact tadpoles spatially avoiding the full predation event cue or tadpole alarm cue may inadvertently encounter a second predator by moving away from the first predator (Fraker and Luttbeg, 2012). Over time, tadpole space use in response to chemical cues may change. Longer-term studies involving repeated chemical cue exposures (such as those by Ferrari and Chivers, 2009; Crane and Ferrari, 2017) would be useful for understanding the role of learning in prey spatial behavior and how it affects predator-prey interactions.

Other studies on the antipredator response of prey also have suggested a longer-term perspective may be necessary. For example prey space use may need to be considered in the context of the physiological stress response (Clinchy et al., 2013). Previous studies on ranid tadpoles have found predator-naive tadpoles exhibit an acute stress response upon predator exposure that is mediated by a reduction in the whole-body concentration of a stress hormone, corticosterone (Middlemis Maher et al., 2013). This acute physiological response permits an adaptive reduction in activity level for a short time. As predator exposure becomes chronic over several days, corticosterone levels rise. This allows an antipredator morphology to develop but also increases the activity level of the tadpoles. Because of the structure of the physiological stress response, spatial avoidance of predators may become more important to a tadpole's (and similar prey's) defense after the stress has become chronic, and especially during the period between the onset of elevated corticosterone and higher activity and the completed development of an antipredator morphology. Focusing on the relative timing of prey learning processes and the physiological stress response in the context of the chemical cue landscape and predator-prey distributions may be productive for understanding how predators and prey interact over space and time.

In contrast to the tadpoles, Anax did not alter their space use patterns in response to any type or combination of chemical cue (Fig. 2B). Although many invertebrates can detect and respond to various chemical cues (Hay, 2009; Kamio and Derby, 2017), we do not know whether Anax cannot detect the tested cues, or whether they can, but did not respond to them. The distribution of Anax still may be influenced indirectly by chemical cues through their interactions with tadpoles that are responding to the cues. However, our results suggest larval Anax use other information sources (e.g., possibly visual information on the distribution of tadpoles, other dragonflies, or habitat structure) to make their space use and movement decisions.

Our results suggest, within the larval dragonfly-tadpole system, chemical cues will affect interactions between predators and prey through the prey response. However, applying these results to natural systems will require further study of how the chemical cue landscape varies. For example physical factors such as advection and dispersion strongly influence the role of chemical cues in predator-prey interactions, especially in more open or dynamic habitats than small shallow ponds (e.g., streams, marine systems; Smee and Weissburg, 2006; Clark and Moore, 2018). Similarly, biological, chemical, and physical environmental factors, including dispersion as well as water chemistry, sunlight and microbial activity, should affect the duration of time that a chemical cue is detectable (i.e., its active time; Peacor, 2006; Wisenden et al, 2009; Chivers et al, 2013; Van Buskirk et al, 2014). Anax will continually release kairomones, but the release of tadpole alarm cues and the kairomone-alarm cue combination may be more discrete events. How long different cue types last in the environment and how tadpoles respond to different cues (including whether or not they learn to recognize risk from kairomones alone) may differ among habitats with different conditions as a result.

Our study provides further information on how chemical cues mediate the spatial interactions between predators and prey. Describing how individuals use the information available to them is a necessary step toward identifying the mechanisms structuring ecological communities.

Acknowledgments.--We thank J. Lebamoff for helping to collect specimens and make experimental observations and members of the Aquatic Ecology Laboratory for helping to revise the manuscript. This research was funded by the Ohio State University Office of Undergraduate Research (Research Scholar Award to TB) and the National Science Foundation (IOS-1557831 to MF and SL).

Literature Cited

Albecker, M. and H. D. Vance-Chalcraft. 2015. Mismatched anti-predator behavioral responses in predator-naive larval anurans. Peer J, 3:e1472.

Anderson, T. L. 2016. Predation risk between cannibalistic aeshnid dragonflies influences their functional response on a larval salamander prey .J. Zool., 300:221-227.

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

Chivers, D. P., D. L. Dixson, J. R. White, M. 1. McCormick, and M. C. O. Ferrari. 2013. Degradation of chemical alarm cues and assessment of risk throughout the day. Ecol. Evol., 3:3925-3934.

Clark, J. and P. A. Moore. 2018. The role of sensory modalities in producing non-consumptive effects for a crayfish-bass predator prey system. Can. J. Zool., in press, cjz-2017-0109

Clinchy, M., M.J. Sheriff, and L. Y. Zanette. 2013. Predator-induced stress and the ecology of fear. Fund. Ecol., 27:56-65.

Crane, A. L. and M. C. O. Ferrari. 2017. Evidence for risk extrapolation in decision making by tadpoles. Sci. Rep., 7:43255.

Ferrari, M. C. O. and D. P. Chivers. 2009. Temporal variability, threat sensitivity and conflicting information about the nature of risk: understanding the dynamics of tadpole antipredator behaviour. Anim. Behav., 78:11-16.

Ferrari, M. C., 15. D. Wisenden, and D. P. Chivers. 2010. Chemical ecology of predator-prey interactions in aquatic ecosystems: a review and prospectus. Can. J. Zool, 88:698-724.

Fraker, M. E. 2008. The dynamics of predation risk assessment: responses of anuran larvae to chemical cues of predators. J. Anim. Ecol., 77:638-645.

Fraker, M. E. and B. Luttbeg. 2012. Predator-prey space use and the spatial distribution of predation events. Behaviour, 149:555-574.

Fraker, M. E., F. Hu, V. Cuddapah, S. A. McCollum, R. A. Relyea, J. Hempel, and R.J. Denver. 2009. Characterization of an alarm pheromone secreted by amphibian tadpoles that induces behavioral inhibition and suppression of the neuroendocrine stress axis. Harm. Behav., 55:520-529.

Friesen, R. G. and D. P. Ciuvers. 2006. Underwater video reveals strong avoidance of alarm cues by fishes. Ethology, 112:339-345.

Gilliam, J. F. and D. F. Fraser. 1987. Habitat selection under predation hazard: test of a model with stream-dwelling minnows. Ecology, 68:1856-1862.

Gosner, K. L. 1960. A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica, 16:183-190.

Hammond, J. I., B. Luttbeg, and A. Sih. 2007. Predator and prey space use: dragonflies and tadpoles in an interactive game. Ecology, 88:1525-1535.

Hay, M. E. 2009. Marine chemical ecology: chemical signals and cues structure marine populations, communities, and ecosystems. Ann. Rev. Mar. Sci., 1:193-212.

Hettyey, A., Z. Toth, K. Thonhauser, J. Frommen, D. Penn, and J. Van Buskirk. 2015. The relative importance of prey-borne and predator-borne chemical cues for inducible antipredator responses in tadpoles. Oecologia, 179:699-710.

Kamio, M. and C. D. Derby. 2017. Finding food: how marine invertebrates use chemical cues to track and select food. Nat. Prod. Rep., 34:514-528.

Laundre. J. W. 2010. Behavioral response races, predator-prey shell games, ecology of fear, and patch use of a large predator and its ungulate prey. Ecology, 91:2995-3007.

Lima, S. L. and T. D. Steury. 2005. Perception of predation risk: the foundation of nonlethal predatorprey interactions, p. 166-188. In: P. Barbosa and I. Castellanos (eds.). Ecology of Predator-Prey Interactions. Oxford University Press.

Lonnstedt, O. M. and M. I. McCormick. 2015. Damsel in distress: captured damselfish prey emit chemical cues that attract secondary predators and improve escape chances. Proc. R. Soc. Lond. Ser. B., 282:2015-2038.

Luttbeg, B. and G. C. Trussell. 2013. How the informational environment shapes how prey estimate predation risk and the resulting indirect effects of predators. Am. Nat., 181:182-194.

Middlemis Maher J., E. E. Werner, and R.J. Denver. 2013. Stress hormones mediate predator-induced phenotypic plasticity in amphibian tadpoles. Proc. R. Soc. Lond. Ser. B., 280:2012-3075.

Mirza, R. S., M. C. O. Ferrari, J. M. Kiesecker, and D. P. Chivers. 2006. Responses of American toad tadpoles to predation cues: behavioural response thresholds, threat-sensitivity and acquired predation recognition. Behaviour, 143:877-889.

Peacor, S. D. 2006. Behavioural responses of bullfrog tadpoles to chemical cues of predation risk is affected by cue age and water source. Hydrobiohgia, 573:39-44.

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

--and E. E. Werner. 1999. Quantifying the relation between predator-induced behavior and growth performance in larval anurans. Ecology, 80:2117-2124.

Schoeppner, N. M. and R. A. Relyea. 2009. Interpreting the smells of predation: how alarm cues and kairomones induce different prey defenses. Fund. Ecol., 23:1114-1121.

Shaffery, H. M. and R. A. Relyea. 2016. Dissecting the smell of fear from conspecific and heterospecific prey: investigating the processes that induce anti-predator defenses Oecologia, 180:55-65.

Sih, A. 2005. Predator-prey space use as an emergent outcome of a behavioral response race. p. 240-255. In: P. Barbosa and I. Castallanos (eds.). Ecology of Predator-Prey Interactions. Oxford University Press.

Smee, D. L. and M. J. Weissburg. 2006. Clamming up: environmental forces diminish the perceptive ability of bivalve prey. Ecology, 87:1587-1598.

Werner, E. E. and S. D. Peacor. 2003. A review of trait-mediated indirect interactions in ecological communities. Ecology, 84:1083-1100.

Van Buskirk. J., A. Krugel, J. Kunz, F. Miss, and A. Stamm. 2014. The rate of degradation of chemical cues indicating predation risk: an experiment and review. Ethology, 120:942-949.

Wisenden, B. D. 2008. Active space of chemical alarm cue in natural fish populations. Behaviour, 145:391-407.

Wisenden, B., M. Rugg, N. Korpi, and L. Fuselier. 2009. Lab and field estimates of active lime of chemical alarm cues of a cyprinid fish and an amphipod crustacean. Behaviour, 146:1423-1442.

Submitted 26 February 2018

Accepted 20 July 2018


Aquatic Ecology Laboratory, Department of Evolution, Ecology, and Organismal Biology, The Ohio State University, Columbus 43212

(1) Corresponding author: e-mail:

Caption: Fig. 1.--.Arena design. Each experimental arena consisted of a 75 L pool separated into four equal quadrants (Q1, Q2, Q3, or Q4). Each quadrant contained an opaque plastic cup providing refuge and an algal disc as a resource patch. Q1, which was randomly determined for each replicate, received the predator or chemical cue treatment

Caption: Fig. 2.--Space use by two size classes of tadpoles (Panel A) and Anax at two feeding states (Panel B) exposed to chemical cues of predation (Treatment abbreviations: K - Anax kairomone, A - tadpole alarm cue). The mean proportion for each treatment is displayed [+ or -] 1 se. A proportion equal to 0.25 indicates that tadpoles or dragonflies were randomly distributed (dashed line). Avoidance is signified by a proportion less than 0.25 and attraction by a proportion greater than 0.25. An asterisk (*) indicates treatments significantly different than the water control
COPYRIGHT 2019 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Brown, Taylor A.; Fraker, Michael E.; Ludsin, Stuart A.
Publication:The American Midland Naturalist
Article Type:Report
Date:Jan 1, 2019
Previous Article:Nest Success and Attributes of Brood Crevices Selected by Green Salamanders (Aneides aeneus) on the Blue Ridge Escarpment.
Next Article:Fish Assemblage Response to Altered Dendritic Connectivity in the Red River Basin, Central Louisiana.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |