Southeastern Naturalist 523 S. Farnsley, B. Kuhajda, A. George, and H. Klug 22001166 SOUTHEASTERN NATURALIST 1V5o(3l.) :1552,3 N–5o3. 33 Fundulus catenatus (Northern Studfish) Response to the Potential Alarm Cue Chondroitin Sulfate Sarah Farnsley1, Bernard Kuhajda2, Anna George2, and Hope Klug1,* Abstract - The evolution of organismal cue-response systems can allow for an effective behavioral reaction to various environmental signals. In aquatic habitats, the reception of certain chemical cues can increase individual fitness of organisms by serving as an indicator of predation threat. In some fish species, damage to an individual’s epidermal cells causes release of a substance that functions as an alarm cue and consequently initiates defense responses in neighboring prey. Recent research on the chemical makeup of the substance that elicits an anti-predator response in Danio rerio (Hamilton) (Zebrafish) revealed that chondroitin fragments were a key component in this substance. The goal of our study was to investigate the ability of chondroitin to elicit an alarm response in Fundulus catenatus (Storer) (Northern Studfish). This species is a small-bodied killifish native to southeastern to south-central USA and is associated with topwater habitats near aquatic and/or overhanging vegetation. We hypothesized that reduced movement and/or a change in position in the water column would be a likely response of the Northern Studfish to chondroitin. We experimentally observed Northern Studfish behavior before and after the addition of chondroitin and a control substance, and compared the fishes’ behavioral responses. Our results show that the Northern Studfish that were exposed to chondroitin tended to reduce their movement by sevenfold and were more likely to move to the bottom of the aquarium relative to the control group, suggesting that chondroitin potentially serves as an alarm-cue component in this species. Our study represents the first demonstration of Northern Studfish response to a chemical cue and the first time that chondroitin sulfate has been shown to elicit a component of alarm behavior in a stream fish. We discuss our findings in relation to potential uses of chondroitin as an alarm cue in the conservation of imperiled stream fishes. Introduction Organisms assess and respond to their environment based on detection of visual, auditory, and/or olfactory cues; specific ecological conditions often dictate the most effective mode of signal transmission (Meuthen et al. 2012). In aquatic environments, chemical cues mediate essential behavioral interactions between individuals when visual cues are limited, e.g., where the degree of turbidity is high or vegetation is dense (Mirza and Chivers 2000). The olfactory reception of chemical cues in aquatic habitats can increase individual fitness by allowing individuals to identify a conspecific mate (Rafferty and Boughman 2006), locate spawning sites (Sorensen et al. 2005), and establish social structure (Moore and Bergman 2005). Chemical 1Department of Biology, Geology and Environmental Science, University of Tennessee-– Chattanooga, Department 2653, 615 McCallie Avenue, Chattanooga, TN 37403. 2Tennessee Aquarium Conservation Institute, 201 Chestnut Street, Chattanooga, TN 37402. *Corresponding author - Hope-Klug@utc.edu. Manuscript Editor: Andrew Rypel Southeastern Naturalist S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 524 cues can also act to decrease the likelihood of predation because various groups of aquatic organisms interpret certain chemical cues as a threat of predation and consequently respond with anti-predator behavior (Ferrari et al. 2010). The odor of a nearby predator (Tollrian and Harvell 1999), the urinary ammonia discharged from startled prey (Kiesecker et al. 1999), and/or substances released following injury inflicted by a predator (Wisenden 2003) have all been shown to serve as alarm signals that initiate defense responses in prey. In many prey species of fishes, such as small-bodied minnows (Cyprinidae) and other members of the superorder Ostariophysi, predatory damage to an individual’s epidermal cells causes release of a substance that functions as a chemical alarm-cue to neighboring conspecifics (Ferrari et al. 2010, Pfeiffer 1977). Originally known as schreckstoff, this alarm substance was first discovered to elicit a fright response in minnows (von Frisch 1941) and has been the focus of several evolutionary, physiological, and chemical studies (Chivers et al. 2007, Mathuru et al. 2012, Pfeiffer et al. 1985). Evidence of response to an alarm cue has now been found in fish species outside the superorder Ostariophysi, including sculpins (Cottidae; Bryer et al. 2000), trout (Salmonidae; Mirza and Chivers 2000), and sunfish (Centrarchidae; Marcus and Brown 2003) (see also Ferrari et al. 2010 for a review). Studies comparing alarm-cue responses between closely related fishes suggest that the alarm cues are likely evolutionarily conserved (Mirza et al. 2003). Chemical alarm-cue recognition is hypothesized to lead to increased survival of neighboring individuals by eliciting an ecologically appropriate defense behavior (Lönnstedt et al. 2012, Mirza and Chivers 2000, Smith 1992). Fishes may respond to alarm-cue detection with area avoidance (Golub et al. 2005), increased shoaling (i.e., grouping together for social reasons, Pitcher 1983), shelter use (Mathis and Smith 1993), altered feeding activity (Mirza and Chivers 2001), and/or complete lack of motion (Reed 1969). A single species may be capable of exhibiting various anti-predator behaviors; the specific response depends on the perceived level of threat (Ferrari et al. 2010). Previous research has focused on the chemical makeup of the cell contents responsible for producing the alarm signal, and some studies have suggested that hypoxanthine-3-N-oxide accounts for an alarm response in Ostariophysan fishes (Brown et al. 2000). Recently, Mathuru et al. (2012) utilized biochemical fractionation to show that chondroitin fragments are a key component of the alarm-substance compound that elicits an anti-predator response in Danio rerio (F. Hamilton) (Zebrafish). Despite a large amount of research on alarm behavior in fishes (Chivers and Smith 1998, Ferrari et al. 2010, Smith 1992), it remains unclear in most species whether chondroitin acts as a component of alarm cues, and if so, what response it elicits. The goal of our study was to investigate the ability of chondroitin to elicit behavior consistent with an alarm response in Fundulus catenatus (Storer) (Northern Studfish), a stream fish widely distributed in the Ohio and Mississippi River drainages (Page and Burr 2011). Although the species that prey upon Northern Studfish in the wild are not well documented, potential predators include Southeastern Naturalist 525 S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 Micropterus dolomieu Lacepède (Smallmouth Bass) and Micropterus salmoides Lacepède (Largemouth Bass) (Tennessee Wildlife Resource Agency 2012). Alarm responses to various chemical stimuli have been well documented in stream fishes. For example, Salvelinus fontinalis (Mitchill) (Brook Trout) respond to damage-released alarm signals of conspecifics as well as alarm cues released from closely related heterospecifics (Mirza and Chivers 2000, 2001), Cottus cognatus Richardson (Slimy Sculpin) respond to predator odor and damage-released alarm cues of conspecifics (Bryer et al. 2001), Etheostoma exile (Girard) (Iowa Dater) respond to epidermal club-cell damage of conspeifics (Smith 1979), and Fundulus olivaceus (Storer) (Blackspotted Topminnow) respond to skin extract from conspecifics (Reed 1969). As such, if chondroitin functions generally as an alarm cue in fishes, we would expect chondroitin to elicit a behavioral response in the Northern Studfish. Other topwater prey-fishes that live near aquatic or overhanging terrestrial vegetation have been shown to cease movement in the presence of predators (Reed 1969), and we therefore hypothesized that reduced movement would be a likely response of the Northern Studfish to this substance. We also hypothesized that Northern Studfish might additionally alter their behavior in response to chondroitin by changing their position in the water column (e.g., by moving near the bottom or top of the water column). To assess these hypotheses, we observed and quantified behavior before and after addition of chondroitin, and compared changes in behavior to those following the addition of a control substance. We discuss our findings in relation to potential uses of the alarm-cue chondroitin in the conservation of stream fishes. Methods Fish acquisition and maintenance This study was approved by the University of Tennessee Chattanooga IACUC (IACUC #:1017HMK-01), and the Tennessee Wildlife Resources Agency (Nashville, TN) issued a scientific collection permit (1691). We used seines to collect adult and juvenile Northern Studfish from the Collins River in McMinnville, TN (35°48'0''N, 85°37'12''W) on 23 May 2013. When not in use in experimental trials, fish were randomly housed together in groups of 8 (regardless of sex, age, or size) in 75.7-L aquaria at the University of Tennessee Chattanooga, Chatanooga, TN. We fed frozen bloodworms to the fish daily, and maintained a 14:10-hour light:dark cycle, and a water temperature of ~18 °C in the aquaria. Experimental trials We conducted our experiments during June and July 2013. To evaluate the effect of the potential alarm-substance chondroitin on Northern Studfish behavior, we performed an experiment that consisted of 2 treatments: (1) exposure to chondroitin and (2) exposure to a control substance (distilled water). Just prior to use in an experimental trial, we measured the total length of each fish and categorized it as either a male, female, or juvenile (Table 1). We began each replicate by randomly selecting 1 fish and placing it in a 75.7-L experimental observation tank, which contained only water and tubing for aeration and substance addition. The observation Southeastern Naturalist S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 526 tank was isolated from all other tanks (i.e., fish in the observation tank did not have any contact with other fish) and from the behavioral observer. Specifically, we made observations from behind an opaque-plastic barrier through eye slits and with the aid of mirrors positioned above the experimental tanks. This set-up allowed us to minimize exposure of the fish to any cues other than the substance added to the tank. Fish were allowed to acclimate for ~1 hour. We observed fish behavior for 10 minutes following acclimation and prior to substance addition. During this time, we recorded movement (i.e., whether the fish was motionless or in motion) and vertical distribution (top, middle, or bottom third) in the tank every 30 sec via scan sampling. After 10 min, we added the experimental substance (either chondroitin or distilled water) through ~5-mm–diameter tubing that had previously been run into the tank. Preliminary studies in which we added dye to the substance confirmed that this method led to the substance being dispersed throughout the tank within seconds. The chondroitin treatment consisted of 0.07 mg of commercially available chondroitin (Sigma C4384, St. Louis, MO) dissolved in 5 mL of distilled water and the control treatment substance was 5 mL of distilled water. This concentration of chondroitin is consistent with levels used in the Mathuru et al. study (2012) that elicited an alarm response in Zebrafish. We conducted a 10-min observation period followed substance addition, during which we assessed the same behaviors as before substance addition. We used each fish only once; a total of 18 trials per treatment were completed (Table 1). Statistical analysis We calculated the proportion of time the fish spent in motion prior to and after substance addition so that we could determine the change in motion for each Table 1. Descriptive data of individuals used in the experiment. We provide an overview of the status of each fish (juvenile, male, or female), the number of fish, the mean ± standard error, and range of the total length of individuals used in each treatment and overall in the study (i.e., in both treatments combined). Treatment Status Number of fish Mean total length ± SE (mm) Range in total length (mm) Control Juvenile 1 44.0 - Male 6 100.7 ± 4.15 85–110 Female 11 81.5 ± 4.85 52–106 Total 18 85.8 ± 4.57 44–110 Chondroitin Juvenile 4 47.5 ± 3.57 42–58 Male 7 103.9 ± 6.30 86–128 Female 7 86.6 ± 7.34 61–117 Total 18 84.6 ± 6.32 42–128 Both combined Juvenile 5 46.8 ± 2.85 42–58 Male 13 102.4 ± 3.77 85–128 Female 18 83.4 ± 4.03 52–117 Total 36 85.2 ± 3.84 42–128 Southeastern Naturalist 527 S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 individual fish following the addition of the substance (chondroitin or distilled water). We also calculated the proportion of time spent on the bottom third of the tank before and after substance addition for each trial to assess the change in use of lower third of the tank for all individuals. We used 2 separate ANCOVA analyses to examine the effect of chondroitin on the change in (1) activity (i.e., the change in the proportion of the time spent in motion) and (2) the change in the proportion of time spent in the bottom third of the aquarium before and after substance addition. Data for both variables were normally distributed (i.e., the data did not differ significantly from a normal distribution; Shapiro-Wilk Test: P = 0.32 and P = 0.24, respectively); thus, the data did not require transformation. We used treatment (chondroitin or distilled water) as a fixed factor in both analyses. Because we hypothesized that size might influence predation risk in the wild and therefore potentially influence the response to chondroitin in our study, we included total length as a covariate in our statistical analyses, and examined the interaction between total length and treatment to assess the relationship between size and behavior for each treatment group. We also examined the relationship between total length and age class and sex using an ANOVA in which status (juvenile, male, female) was a fixed factor and body length was the response variable. Body length was normally distributed (Shapiro- Wilk Test: P = 0.11) and did not require transformation prior to use in this analysis. We employed a Tukey’s HSD test to detect any significant differences among treatment groups. Results Chondroitin significantly decreased motion in the Northern Studfish (treatment effect: F1, 32 = 10.1, P = 0.003; Fig. 1). Following substance addition, fish that were exposed to chondroitin tended to reduce their movement by sevenfold relative to the control group, with change in proportion of time spent in motion as -0.354 versus -0.050 (see also Table 2 for additional details of time spent in motion across treatments before and after substance addition). Overall, there was no significant relationship between total length and the change in movement (F1, 32 = 3.90, P = 0.057). However, there was a significant interaction between treatment and total length (F1, 32 = 5.92, P = 0.021). Specifically, in our control group, only relatively long fish tended to decrease movement after the addition of our control substance (water) (Fig. 2A), whereas the majority of fish, regardless of size, reduced movement after the addition of chondroitin (Fig. 2B). Fish exposed to chondroitin were significantly more likely than fish in the control group to increase the proportion of time spent on the bottom third of the tank following substance addition (treatment effect: F1, 32 = 4.37, P = 0.045; Table 2, Fig. 3). However, there was no relationship between length and the change in the use of the bottom third of the tank (F1, 32 = 0.10, P = 0.76) and no significant interaction between treatment and total length (F1, 32 = 3.27, P = 0.08). Southeastern Naturalist S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 528 There was a significant relationship between the total length of a fish and whether that fish was a juvenile, male, or female (F1, 33 = 25.3, P < 0.001). Specifically, juveniles were significantly shorter than both males (Tukey’s HSD: mean difference = -55.6, P < 0.001) and females (Tukey’s HSD: mean difference = -36.6, P less than 0.001), and females were significantly shorter than males (Tukey’s HSD: mean difference = -18.9, P = 0.004; Table 2). Given this result, the relationship between body length, treatment, and activity described above was reflective of maturation (i.e., juvenile versus adult) or sex dif ferences among individuals. Table 2. Mean ± standard error and the range of the proportion of time spent in motion and in the bottom third of the aquarium before and after the addition of distilled water (control treatment) and chondroitin mixed with distilled water. n = 18 for both treatment groups. Proportion of time in motion Proportion of time in bottom third of tank Treatment Before addition After addition Before addition After addition Control Mean ± SE 0.380 ± 0.079 0.320 ± 0.091 0.630 ± 0.079 0.660± 0.092 Range 0.000–1.000 0.000–1.000 0.095–1.000 0.048–1.000 Chondroitin Mean ± SE 0.670 ± 0.084 0.310 ± 0.80 0.440 ± 0.079 0.620 ± 0.078 Range 0.000–1.000 0.000–1.000 0.000–1.000 0.000–1.000 Figure 1. Mean change in the proportion of time spent in motion before substance addition and after substance addition for control (distilled water) and chondroitin treatment. Whiskers represent ± standard error. Southeastern Naturalist 529 S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 Figure 2. The relationship between total length and the change in the proportion of time spent in motion before substance addition and after substance addition for (A) control (distilled water) and (B) chondroitin treatments. There was a significant relationship between total length and the change in the proportion of time spent active for (A) control fish (linear regression: r2 = 0.36, P = 0.009) but not for (B) fish exposed to chondroitin (linear regression: r2 = 0.008, P = 0.72). Southeastern Naturalist S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 530 Discussion Chondroitin significantly reduced the time Northern Studfish spent in motion and caused fish to spend a greater proportion of time near the bottom of the aquarium than those fish without exposure to chondroitin. Further, in contrast to the case in which only water was added to the tank, nearly all fish, regardless of size, reduced activity when chondroitin was added to the aquarium. Together, these findings suggest that chondroitin potentially serves as a component of alarm cues in this species. Northern Studfish live in or near aquatic or overhanging terrestrial vegetation; thus, decreasing motion and moving towards the bottom of the water column are likely adaptive responses to a predation threat (Reed 1969). Ours is the first study to demonstrate Northern Studfish response to a chemical cue, and it is the first time that chondroitin sulfate has been shown to elicit behavior consistent with an alarm response in a stream fish. In the future, additional work that compares the behavioral response elicited by chondroitin with the behavioral response elicited by the epidermal cells of con- and/or hetero-specific fish would be useful and would provide a more-complete picture of how chondroitin potentially functions as an alarm cue. The finding that chondroitin potentially functions as an alarm cue is consistent with work by Mathuru et al. (2012), who found that chondroitin fragments trigger a fear response in Zebrafish. Likewise, the finding that the Northern Studfish exhibits a form of anti-predator behavior in response to a chemical cue is consistent with Figure 3. Mean change in the proportion of time spent in the bottom one-third of the aquarium before substance addition and after substance addition for control (distilled water) and chondroitin treatment. Whiskers represent ± standard error. Southeastern Naturalist 531 S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 studies of other stream fishes. For example, Bryer et al. (2001) found that Slimy Sculpins increased shelter use in response to damage-released cues from conspecifics and that the presence of these cues intensified alarm response to predator odor. Brook Trout have been shown to respond to alarm signals released upon injury to conspecifics by decreasing activity and reducing foraging (Mirza and Chivers 2001). Interestingly, we also found that a change in the amount of movement of fish in the control group following the treatment was correlated with body size but the same was not true for fish in the group that was exposed to chondroitin. Specifically, fish of all sizes exposed to chondroitin tended to reduce activity and to the same degree. It is possible that in the absence of predation threat, larger fish naturally exhibit a greater response to disturbance, and in nature, this might be due to differences in feeding or territorial behavior. However, our finding that individuals of all sizes reduced activity when exposed to chondroitin further suggests that chondroitin elicits anti-predator behavior. Our results have implications for the conservation of imperiled fishes, many of which are stream-dwelling species. We hypothesize that by pairing a predatory cue with a potential alarm substance, captive or hatchery-raised fish can be conditioned to associate a predator with an anti-predator response. Once released back into their natural habitat, survival of these fish may increase as a result of this associative learning. Mirza and Chivers (2000) showed increased survival among juvenile Brook Trout conditioned to associate a damage-released alarm cue with predator odor. These conditioned fish retained predator recognition in response to predator odor for 10 days following conditioning (Mirza and Chivers 2000). Other species for which staged predator-encounters led to increased survival based on past experience with predator cues include juvenile Oncorhynchus mykiss (Walbaum) (Rainbow Trout; Mirza and Chivers 2003) and Pimephales promelas Rafinesque (Fathead Minnow; Gazdewich and Chivers 2002). Predator-recognition conditioning is particularly useful for conservation when predators are an unfamiliar, introduced species and represent a considerable threat to prey-species survival. Following conditioning, prey are no longer naïve to a predator and are better equipped to assess risk of predation and exhibit ef fective defensive behaviors. Future studies should aim to determine whether threatened and endangered fishes respond to chondroitin as an alarm cue, and, if so, whether the use of this chemical cue can be used in conditioning those fishes to associate it with a predation threat. Acknowledgments We are grateful to Kathlina Alford and David Neely for their advice on fish acquisition and care. We also thank the editor and 2 anonymous reviewers for their comments, which greatly improved the manuscript. Literature Cited Brown, G.E., J.C. Adrian, E. Smith, H. Leet, and S. Brennan. 2000. Ostariophysan alarmpheromones: Laboratory and field-test of the functional significance of nitrogen oxides. Journal of Chemical Ecology 26:139–154. Southeastern Naturalist S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 532 Bryer, P., R.S. Mirza, and D.P. Chivers. 2001. Chemosensory assessment of predation risk by Slimy Sculpins (Cottus cognatus): Responses to alarm, disturbance, and predator cues. Journal of Chemical Ecology 27:533–546. Chivers, D.P., and R.J.F. Smith. 1998. Chemical-alarm signaling in aquatic predator–prey systems: A review and prospectus. EcoScience 5:338–352. Chivers, D.P., B.S. Wisenden, C.J. Hindman, T. Michalak, R.C. Kusch, S.W. Kaminskyj, K.L. Jack, M.C.O. Ferrari, R.J. Pollock, C.F. Halbgewachs, M.S. Pollock, S. Alemadi, C.T. James, R.K. Savaloja, C.P. Goater, A. Corwin, R.S. Mirza, J.M. Kiesecker, G.E. Brown, J.C. Adrian Jr., P.H. Krone, A.R. Blaustein, and A. Mathis. 2007. Epidermal “alarm” substance cells of fishes maintained by non-alarm functions: Possible defense against pathogens, parasites, and UVB radiation. Proceedings of the Royal Society of London B 274:2611–2619. Ferrari, M.C.O., B.D. Wisenden, and D.P. Chivers. 2010. Chemical ecology of predator– prey interactions in aquatic ecosystems: A review and prospectus. Canadian Journal of Zoology 82:698–724. Gazdewich, K.J., and D.P. Chivers. 2002. Acquired predator-recognition by Fathead Minnows: Influence of habitat characteristics on survival. Journal of Chemical Ecology 28:439–445. Golub, J.L., V. Vermette, and G.E. Brown. 2005. Response to conspecific and heterospecific alarm cues by Pumpkinseeds in simple and complex habitats: Field verification of an ontogenetic shift. Journal of Fish Biology 66:1073–1081. Kiesecker, J.M., D.P. Chivers, A. Marco, C. Quilchano, M.T. Anderson, and A.R. Blaustein. 1999. Identification of a disturbance signal in Red-legged Frogs, Rana aurora. Animal Behaviour 57:1295–1300. Lönnstedt, O.M., M.I. McCormick, M.G. Meekan, M.C.O. Ferrari, and D.P. Chivers. 2012. Learn and live: Predator experience and feeding history determines prey behavior and survival. Proceedings of the Royal Society of London B 279:2091 –2098. Marcus, J.P., and G.E. Brown. 2003. Response of Pumpkinseed Sunfish to conspecific chemical alarm cues: An interaction between ontogeny and stimulus concentrations. Canadian Journal of Zoology 81:1671–1677. Mathis, A., and R.J.F. Smith. 1993. Chemical-alarm signals increase the survival time of Fathead Minnows (Pimephales promelas) during encounters with Northern Pike (Esox lucius). Behavioral Ecology 4:260–265. Mathuru, A.S., C. Kibat, W.F. Cheong, G. Shui, M.R. Wenk, R.W. Friedrich, and S. Jesuthasan. 2012. Chondroitin fragments are odorants that trigger fear behavior in fish. Current Biology 22:538–544. Meuthen, D., S.A. Baldauf, and T. Thünken. 2012. Evolution of alarm cues: A role for kin selection? F1000 Research. 1:1–27. Mirza, R.S., and D.P. Chivers. 2000. Predator-recognition training enhances survival of Brook Trout: Evidence from laboratory and field-enclosure studies. Canadian Journal of Zoology 78:2198–2208. Mirza, R.S., and D.P. Chivers. 2001. Are chemical-alarm cues conserved within salmonid fishes? Journal of Chemical Ecology 27:1641–1655. Mirza, R.S., and D.P. Chivers. 2003. Response of juvenile Rainbow Trout to varying concentrations of chemical-alarm cue: Response thresholds and survival during encounters with predators. Canadian Journal of Zoology 81:88–95. Mirza, R.S., S.A. Fisher, and D.P. Chivers. 2003. Assessment of predation risk by juvenile Yellow Perch, Perca flavescens: Response to alarm cues from conspecifics and preyguild members. Environmental Biology of Fishes 66:321–327. Southeastern Naturalist 533 S. Farnsley, B. Kuhajda, A. George, and H. Klug 2016 Vol. 15, No. 3 Moore, P.A., and D.A. Bergman. 2005. The smell of success and failure: The role of intrinsic and extrinsic chemical signals on the social behavior of crayfish. Integrative and Comparative Biology 45:650–657. Page, L.M., and B.M. Burr. 2011. Peterson Field Guide to Freshwater Fishes of North America North of Mexico. 2nd Edition. Houghton-Mifflin-Harcourt Publishing, New York, NY. 688 pp. Pfeiffer, W. 1977. Distribution of fright reaction and alarm-substance cells in fishes. Copeia 4:653–665. Pfeiffer, W., G. Riegelbauer, G. Meier, and B. Scheibler. 1985. Effect of hypoxanthine- 3(N)-oxide and hypoxanthine-1(N)-oxide on central-nervous excitation of the Black Tetra Gymnocorymbus ternetzi (Characidae, Ostariophysi, Pisces) indicated by dorsallight response. Journal of Chemical Ecology 11:507–523. Pitcher, T.J. 1983. Heuristic definitions of shoaling behavior. Animal Behavior. 31:611–613. Rafferty, N.E., and J.W. Boughman. 2006. Olfactory mate-recognition in a sympatric species pair of Three-spined Sticklebacks. Behavioral Ecology 17:965–970. Reed, J.R. 1969. Alarm substances and fright reaction in some fishes from the Southeastern United States. Transactions of the American Fisheries Society 4:664–668. Smith, R.J.F. 1979. Alarm reaction of Iowa and Johnny Darters (Etheostoma, Percidae, Pisces) to chemicals from injured conspecific. Canadian Journal of Zoology 57:1278–1282. Smith, R.J.F. 1992. Alarm signals in fishes. Reviews in Fish Biology and Fisheries 2:33–63. Sorensen, P.W., J.M. Fine, V. Dvornikovs, C.S. Jeffrey, F. Shao, J. Wang, L.A. Vrieze, K.R. Anderson, and T.R. Hoye. 2005. Mixture of new sulfated steroids functions as a migratory pheromone in the Sea Lamprey. Nature Chemical Biology 1:324–328. Tennessee Wildlife Resources Agency. 2012. The angler’s guide to Tennessee fish. Available online at https://www.tn.gov/assets/entities/twra/fish/attachments/anglersguide. pdf. Accessed 5 December 2013. Tollrian R., and C.D. Harvell. 1999. The Evolution of Inducible Defenses. Princeton University Press, Princeton, NJ. 383 pp. von Frisch, K. 1941. Uber einen Schreckstoff der Fischhaut und seine biologische Bedeutung. S. Zeitschrift für vergleichende Physiologie 29:46–145. Wisenden, B.D. 2003. Chemically mediated strategies to counter predation. Pp. 236–251, In S.P. Collin and N.J. Marshall (Eds.). Sensory Processing in Aquatic Eenvironments. Springer Verlag, New York, NY. 446 pp.