Chemo-268.qxd

2/24/04 3:20 PM

Page 107

Chemoecology 14:107–112 (2004) 0937–7409/04/020107–6 © Birkhäuser Verlag, Basel, 2004 DOI 10.1007/s00049-003-0268-7

CHEMOECOLOGY

Learning to recognize novel predators under weakly acidic conditions: the effects of reduced pH on acquired predator recognition by juvenile rainbow trout Antoine O.H.C. Leduc, Maud C.O. Ferrari, Jocelyn M. Kelly and Grant E. Brown Department of Biology, Concordia University, 7141 Sherbrooke St. W., Montréal, Québec, H4B 1R6, Canada

Summary. Recent studies have demonstrated that under weakly acidic conditions (pH 6.0), many prey fishes, including juvenile rainbow trout (Onchorhynchus mykiss), do not exhibit overt antipredator responses to conspecific chemical alarm cues. In laboratory trials, we investigated the potential effects of reduced pH on the ability of hatchery reared, predator naïve juvenile rainbow trout to acquire the recognition of a novel predator (yellow perch, Perca flavenscens). Initially, we exposed trout to the odour of a predatory yellow perch, buffered to pH 6.0 (weakly acidic) or pH 7.0 (neutral) paired with conspecific skin extracts (also buffered to pH 6.0 or 7.0) or a distilled water control. Juvenile trout exhibited significant increase in antipredator behaviour when exposed to neutral skin extract (pH 7.0). When retested 48 hours later to perch odour alone (pH 7.0), only trout initially conditioned with neutral skin extracts (pairs with either neutral or acidic perch odour) exhibited a learned recognition of perch odour as a predator risk. Those initially exposed to weakly acidic skin extract or the distilled water control did not show a learned response to predator odour. These results demonstrate that the ability to acquire the recognition of novel predators is impaired under weakly acidic conditions, as would occur in natural waterways affected by acidic precipitation. Key words. Chemical alarm cues – salmonids – acid rain – antipredator behaviour – acquired predator recognition – rainbow trout

Introduction Recently, a large volume of research has focused on the impact of acid rain on aquatic ecosystems (Atland 1998; Barry et al. 2000; Hesthagen & Jonsson 2002). Sulphur and nitrogen oxides, released during fossil fuel combustion, are a major source of water acidification worldwide (Rodhe et al. 1995), and result in acidification ranging from weakly acidic (pH 6.0-6.9) to heavily acidic (pH < 5.0) (Rodhe et al. 1995). Much is known regarding the effects of heavily acidic conditions on aquatic communities (e.g., Somers & Harvey 1984; Haines & Baker 1986; Baker et al. 1990). The major impacts of heavily acidified waters (pH < 5.0) include Correspondence to: Grant E. Brown, e-mail: [email protected]

decreases in fish abundance and recruitment, and increases in physiological stress and mortality (Schindler et al. 1985; Environment Canada 1997). Juveniles exposed to intermediate pH (∼5.0 to 6.0) show lower growth rate, reduced swimming performances, impairment of sensory mechanisms and reduced immunity (Lorz & McPherson 1976). Despite the extensive research conducted on this topic, we have a poor understanding of the effects of weakly acidic conditions (pH 6.0-6.9) on prey fish behaviour, especially, the use of chemical alarm cues to mediate predation risk. Damage-released chemical alarm cues have been shown in a wide variety of fishes, including ostariophysans, salmonids, gobies, poeciliids, sticklebacks, percids, sculpins, cottids, cichlids, and centrarchids (Chivers & Smith 1998; Brown 2003). When detected by nearby conspecifics and some sympatric heterospecifics, typically following mechanical damage to the skin (as would occur during a predation attempt), these chemical alarm cues elicit dramatic short-term increase in species-typical antipredator behavioural response. Such overt behavioural responses (Smith 1999) have been well documented and include a variety of behaviour patterns, such as increased shoal cohesion, area avoidance, dashing and freezing, and decreased foraging and mating (Chivers & Smith 1998). Responding to both conspecific and heterospecific alarm cues has been shown to increase the probability of survival during predation events (Mathis & Smith 1993; Chivers et al. 1996; Mirza & Chivers 2000, 2003). Thus, any impairment in the ability to detect and respond to alarm cues would likely have significant fitness costs (Brown et al. 2002). Chemical alarm signals can also elicit a variety of covert behavioural responses (Smith 1999) including acquired recognition of novel predators, morphological and life history changes and acquired information regarding local predation risk through predator inspection behaviour (Smith 1999; Brown 2003). One of the most widely studied examples of covert responses to the detection of chemical alarm cues is the learned recognition of predator identities (Chivers & Smith 1998; Smith 1999). Many prey fishes do not show innate recognition of potential predators. Rather, they must acquire this knowledge based on the pairing of alarm cues with the visual and/or chemical cues of the predator (Chivers & Smith 1998; Smith 1999). For example, European minnows (Phoxinus phoxinus) and fathead minnows

Chemo-268.qxd

108

2/24/04 3:20 PM

Page 108

A.O.H.C. Leduc et al.

(Pimephales promelas) acquire the recognition of the chemical cues (odour) of a novel predator after a single exposure to the predator cues paired with conspecific alarm cues (Magurran 1989; Mathis & Smith 1993). Fathead minnows similarly learn to recognize the visual cues of a predator following a single conditioning experience (Chivers & Smith 1994). The learned recognition of novel predators increases an individual’s survival during subsequent predator encounters (Mirza & Chivers 2000, 2001; Gazedwich & Chivers 2002). The ability to detect and/or respond to these critically important chemical alarm cues is impaired under weakly acidic conditions (pH 6.0) in at least three groups of prey fishes. When exposed to conspecific skin extract or hypoxanthine-3-N-oxide (H3NO; the putative ostariophysan alarm pheromone, Brown et al. 2000, 2001a, 2003; Brown 2003), under weakly acidic conditions, two cyprinid species (fathead minnows and finescale dace, Phoxinus neogaeus) did not exhibit a species-typical increase in antipredator behaviour (Brown et al. 2002). This impairment appears to be due to a non-reversible structural change in the alarm molecule itself, and not due to damage of the olfactory receptors (Brown et al. 2002). A similar loss of response in weakly acidic conditions has been demonstrated in juvenile pumpkinseed sunfish (Lepomis gibbosus). Under weakly acidic conditions (∼pH 6.0), sunfish exhibited significantly weaker antipredator responses to conspecific alarm cues and those of an allopatric congener (green sunfish, Lepomis cyanellus) and a total loss of response to a cyprinid alarm cue (Leduc et al. 2003). Finally, in laboratory and field trials, two species of juvenile salmonids (rainbow trout, Oncorhynchus mykiss and brook charr, Salvelinus fontinalis) did not respond to conspecific skin extracts at pH 6.0-6.2 (Leduc et al., in press). The absence of an overt antipredator response, however, does not mean that prey cannot detect the cues. Fathead minnows, for example, are able to learn to recognize the chemical cues of a novel predator (yellow perch, Perca flavescens) when the predator cues are paired with alarm cues well below the concentration required to elicit an overt antipredator response (Brown et al. 2001b). Similarly, juvenile rainbow trout successfully avoid predatory northern pike (Esox lucius) when exposed to trout alarm cues well below the minimum behavioural response threshold (Mirza & Chivers 2003). This demonstrates that individuals can detect conspecific alarm cues, and acquire the recognition of novel predators well below the concentration needed to elicit an overt behavioural response. It remains unknown, however, how a relatively minor shift in pH results in the loss of function of chemical alarm cues in salmonid fishes. Two alternative hypotheses are possible. Initially, a decrease in pH may completely degrade the alarm cues, as seen in cyprinids (Brown et al. 2002). Alternatively, a decrease in pH may reduce the functional concentration of the alarm cue to a level near the minimum behavioural response threshold, as seen for centrarchids (Leduc et al. 2003). To test these alternatives, we exposed juvenile rainbow trout to paired conspecific skin extracts (neutral [pH 7.0] vs. weakly acidic [pH 6.0]) and the odour of a novel predator (neutral vs. weakly acidic). Later, we tested for the acquired recognition of the predator odour in

CHEMOECOLOGY

order to assess which of these alternative hypotheses was correct. If decreased pH results in complete degradation of the alarm cues, we predict no learned recognition of novel predator odour. However, if the weakly acidic conditions results in only a partial degradation of the alarm cues, we predict that juvenile trout should still be able to acquire the recognition of the novel predator.

Methods Test fish Juvenile rainbow trout, obtained from a local hatchery, were housed in 390-L tank of dechlorinated tap water at 14°C and fed ad libitum twice daily with commercial trout food. The photoperiod was adjusted to 14:10 L:D cycles. Adult yellow perch were captured from the Lachine canal, near Montreal, Quebec, Canada housed in a 390-L tank with dechlorinated tap water at 18-20°C, fed ad libitum with brine shrimp (Artemia spp) and Tetramin cichlid pellets. Swordtails (Xiphophorus helleri) were obtained from a commercial supplier, fed with commercial flakes and held in a 110-L aquarium at approximately 26°C. Stimulus collection Trout skin extract (SE): We collected conspecific skin extract from 12 juvenile rainbow trout (mean ± S.E. fork length = 5.95 ± 0.25 cm). Trout were killed with a blow on the head (in accordance with Concordia University Animal Care Committee protocol #AC2002-BROW), and skin fillets were removed from either side of the body and placed in 50 mL of chilled distilled water. Skin fillets were then homogenized and filtered through glass wool (to remove any remaining tissues) and diluted to the desired final concentration. We collected a total of 79.5 cm2 of skin (in 895 mL of distilled water). The concentration used during the experiment was the same as used for previous studies (Brown & Smith 1997, 1998). Skin extracts were frozen in 30 mL aliquots at –20°C until required. As a control, we also froze 30 mL aliquots of glass-distilled water. Perch odour (PO): Prey animal often exhibit antipredator responses to chemical cues of predators fed conspecifics of the prey, but not to those of predators fed another diet (Chivers & Mirza 2001). Consequently, we fed two adults yellow perch (12.8 and 14.2 cm standard length) with swordtails prior to stimulus collection. Perch were fed one green swordtail daily during four consecutive days. We used swordtails as a control diet, since they are phylogenetically distant and allopatric from rainbow trout, and lack an alarm cue recognized by rainbow trout (Brown & Smith 1997). One hour after the final feeding, we transferred each perch to a 5 L stimulus-collection tank (filled with 3 L of dechlorinated tap water). The stimulus-collection tank water was aerated but not filtered. The perch remained in the stimulus-collection tank for 48 h, at which time they were removed and returned to their holding tank. We pooled that water from both tanks and filtered it through filter floss to remove any particulate matter. Water containing perch odour was frozen at –20°C in 50 mL aliquots until required. Test stimuli Trout were exposed to conspecific skin extract and perch odour, which had been buffered to pH 6.0 with the addition of approximately 0.1 mL of dilute H2SO4 or left untreated (pH 7.0-7.2). Sulphuric acid was used to lower pH as it is known to be a major source of surface water acidification (Rodhe et al. 1995). We buffered the stimulus directly rather than the test tank, since Brown et al. (2002) have shown no significant difference in the impairment of the antipredator response of fathead minnow between acidifying their environment (water) and the experimental stimulus itself (skin extract).

Chemo-268.qxd

2/24/04 3:20 PM

Page 109

Vol. 14, 2004 Test tank The test tank consisted of two stream-channels (2.50 × 0.66 m), equipped with a single chiller unit set to 12°C and filled with dechlorinated tap water. A submersible aquarium pump was used to generate a water current of approximately 5 cm sec−1. The two stream channels have separated inflow and outflow, eliminating possible mixing of water between them. Mesh screens were fixed at either end of the channel, confining the fish to a section 1.50 m long. A single shelter consisting in a ceramic tile (16 × 16 cm) mounted on four cylinders (6 cm height) was placed in the middle of each channel, delimitating a front area. Stimulus injection tubes were placed at the upstream end of the channel, allowing for the injection of experimental stimulus without disturbing the test fish. Conditioning trials The experiment consisted of two phases: conditioning trials, followed by recognition trials. During conditioning trials, trout were exposed to a paired stimulus of conspecific skin extract (at pH 6.0 or 7.0) and perch odour (at pH 6.0 or 7.0) or a control stimulus of distilled water paired with perch odour (pH 7.0). Prior to testing, two juvenile rainbow trout were acclimated for a period of 24 hours in a test stream channel, to compensate for apparent stress (personal observations). However, only one trout, chosen randomly, was used for subsequent behavioural measurements. Trout were placed in the same photoperiod as mentioned above. We used 100 trout (fork length: mean ± SE = 5.57 ± 0.10 cm; weight: mean ± SE = 3.11 ± 0.16 g), N = 20 per treatment. Observation consisted of a 10 minute pre-stimulus and a 10 minute post-stimulus observation periods. Prior to the pre-stimulus observation, we withdrew and discarded 60 mL of water from the injected tubes (to remove any stagnant water) and withdrew and retained an additional 60 mL. Following the pre-stimulus observation period, we injected one of five stimulus combinations: (1) 20 mL of SE(pH7) and 20 mL of PO(pH7), (2) 20 mL of SE(pH7) and 20 mL of PO(pH6), (3) 20 mL of SE(pH6) and 20 mL of PO(pH7). (4) 20 mL of SE(pH6) and 20 mL of PO(pH6), or (5) 20 mL of distilled water and 20 mL of PO(pH7). We used the retained tank water to slowly flush the stimuli into the tank. Once the stimulus was fully injected, we began the post-stimulus observation period. All behavioural observations were filmed using a remotely controlled black and white low-light “Sea-Viewer” camera. At all time, the experimenter was out of the range of sight of test fish so as not to disturb their response. From the videotapes, we recorded three behavioural measures: time spent moving, time spent under cover and time spend in front area (where the stimuli were delivered). Reduced movement, increasing shelter use, and increased avoidance of the area where the cues have been detected are typical antipredator response (Brown & Smith 1997; Brown 2003). Statistical analysis For each behavioural measure, we calculated the difference between pre- and post-stimulus observation periods, and used these differences values as dependent measures in all subsequent analyses. As the three behavioural measures are correlated, we initially tested for differences between stimulus combinations using a MANOVA. Subsequent ANOVAs were used to examine differences within each behavioural measure. Post-hoc comparisons were made using Fisher’s Protected Least Squared Differences. Recognition trials After the conditioning trials, fish were transferred in a second, identical stream channel, fed ad libitum for two days and then retested for the learned recognition of predator odour. The recognition trials followed the same experimental protocol as described above, except they were exposed to 20 mL of perch odour alone (at pH 7.0). We predicted that if individuals had acquired the recognition of the predator cue, perch odour alone would elicit a significant

Acquired predator recognition under weakly acidic conditions

109

increase in antipredator behaviour. Data were analysed as described above.

Results Conditioning trials

We found a significant overall difference among treatment combinations (F(4,45) = 25.49, P < 0.0001). Trout exposed to skin extract at pH 7.0, regardless of perch odour pH, exhibited a significant decrease in time spent moving (F(4, 45) = 16.75, P = 0.0001, Fig. 1A), and time in front area (F(4,45) = 6.79, P = 0.0002; Fig. 1B) and a significant increase in time under shelter (F(4,45) = 4.99, P = 0.002; Fig. 1C). Those exposed to skin extract at pH 6.0 were not significantly different from the distilled water paired with neutral perch odour control (Fig. 1). Recognition trials

As with the conditioning trials, we found a significant overall difference among treatment combinations (i.e. original conditioning treatments) (F(4,45) = 7.42, P = 0.0001). When exposed to perch odour alone, trout initially exposed to skin extract at pH 7.0, regardless of perch odour pH, exhibited a significant decrease in time spent moving (F(4,45) = 5.79, P = 0.0007; Fig. 2A), and time in the front area (F(4,45) = 4.79, P = 0.0021; Fig. 2B) and a significant increase in time under shelter (F(4,45) = 5.31, P = 0.0014; Fig. 2C). Those initially exposed to skin extract at pH 6.0 were not significantly different from the control group initially exposed to distilled water paired with neutral perch odour (Fig. 2).

Discussion These data demonstrate that the ability to acquire the recognitions of a novel predator is impaired under weakly acidic conditions. During the conditioning trials, only trout exposed to neutral (pH 7.0) skin extracts exhibited a significant increase in antipredator behaviour. Those exposed to weakly acidic (pH 6.0) skin extracts did not exhibit any changes in behaviour (i.e. were not different from the distilled water controls). When exposed to perch odour alone, trout initially exposed to neutral pH skin extracts exhibited a learned antipredator response, while those exposed to weakly acidic skin extracts did not. Moreover, the data suggest that the pH of the predator odour has no effect on either the antipredator response or learned recognition of the predator odour. While previous studies have demonstrated a loss of overt response due to the effects of weakly acidic conditions in prey fishes (Brown et al. 2002; Leduc et al. 2003, in press), this is the first demonstration that weakly acidic conditions can also lead to the loss of covert response, such as acquired predator recognition. The observed loss of response to conspecific chemical alarm cues (Brown et al. 2002; Leduc et al. 2003; Leduc et al., in press) may be due to a complete or only a partial degradation of the chemical signal. Previous work has

Chemo-268.qxd

110

2/24/04 3:20 PM

A.O.H.C. Leduc et al. 100

CHEMOECOLOGY

A

100

b

50

b

a

Time moving (sec)

−100 −150 −200 −250

b

−50 −100 −150 −200

−350

SE7-PO7 SE7-PO6 SE6-PO7 SE6-PO6 DW-PO7 B

SE7-PO7 SE7-PO6 SE6-PO7 SE6-PO6 DW-PO7 100

b b a

0

b

B

50

b

a

a

0 Time in front area (sec)

Time in front area (sec)

b

−300

50

−50 −100 −150 −200

b

−100 −150 −200 −250

−300

−300

−350

a

b

−50

−250

−350

SE7-PO7 SE7-PO6 SE6-PO7 SE6-PO6 DW-PO7 200

a

−250

−300

100

a

0

−50

−350

b

A

50

b a

0 Time moving (sec)

Page 110

SE7-PO7 SE7-PO6 SE6-PO7 SE6-PO6 DW-PO7 200

C

C

a 150

a

100

50 b

b

b

0

−50

Time under shelter (sec)

Time under shelter (sec)

150

a a

100

50

0

b

b

b

−50

−100 SE7-PO7 SE7-PO6 SE6-PO7 SE6-PO6 DW-PO7

Fig. 1 Mean (± S.E.) change in time (in seconds) spent moving (A), time in front (stimulus delivery) area of the test tank (B), and time under shelter (C) for juvenile rainbow trout exposed to each of the treatment combinations during the conditioning trial. SE7-PO7= neutral (pH 7.0) skin extract paired with neutral perch odour; SE7PO6 = neutral skin extract paired with weakly acidic (pH 6.0) perch odour; SE6-PO7= weakly acidic skin extract paired with weakly acidic perch odour; DW-PO7= distilled water paired with neutral perch odour. Bars under different letters denote significant differences at P < 0.05, based on Fisher’s Protected Least Squared Differences

−100

SE7-PO7 SE7-PO6 SE6-PO7 SE6-PO6 DW-PO7

Fig. 2 Mean (± S.E.) change in time (in seconds) spent moving (A), time in front (stimulus delivery) area of the test tank (B), and time under shelter (C) for juvenile rainbow trout exposed to perch odour alone during the learned recognition trials. Stimulus labels (as in Fig. 1) on x-axis refer to treatment combinations used during initial conditioning trials (see text). Bars under different letters denote significant differences at P < 0.05, based on Fisher’s Protected Least Squares Differences

Chemo-268.qxd

2/24/04 3:20 PM

Page 111

Vol. 14, 2004

shown that the cyprinid alarm pheromone is completely degraded under weakly acidic conditions (Brown et al. 2002). The centrarchid alarm cue, however, appears to be only partially degraded under similar weakly acidic conditions (Leduc et al. 2003). Pumpkinseed sunfish exhibited a weak, yet consistent response to conspecific alarm cues at pH 6.0, but no response to cyprinid alarm cue (Leduc et al. 2003). If the salmonid alarm cues were only being partially degraded, we would have expected to see some evidence of learned recognition. Thus, these data suggest that under weakly acidic conditions, the salmonid alarm cues are completely degraded. Throughout Europe and North America, the number of salmonids released from hatcheries for restocking purpose now matches or exceeds natural production (Brown & Laland 2001). It is known that high levels of mortality can be attributed to naïve hatchery-reared individuals failing to recognize predators due to an inability to acquire appropriate cues prior to stocking (Suboski & Templeton 1989). As a result, newly stocked individuals are at higher risk of predation compared to wild strains (Donnelly & Whoriskey 1993; Shiverly et al. 1996; Brown & Laland 2001). Mirza & Chivers (2000) have demonstrated that acquired predator recognition significantly increases survival of brook charr. Individuals conditioned to recognize the chemical cues of chain pickerel (Esox niger) as a predation threat were better able to evade the predator than non-trained fish. Thus, these results combined with our data strongly suggest the inability to detect conspecific alarm cues and to use these cues to acquire the recognition of novel predator could have significant fitness costs (i.e. increased predation risk) under weakly acidic conditions. Many prey species rely on heterospecific alarm cues for the assessment of local predation risk and acquired predator recognition (Smith 1999; Brown 2003). In addition, heterospecific chemical cues have also been demonstrated to function as foraging cues (Smith 1999; Golub & Brown 2003). Individuals may also acquire the recognition of novel predators based on the presence of heterospecific alarm cues (Brown 2003). Juvenile salmonids are often found coexisting with a variety of heterospecifics, including other salmonids (Raffenerg & Parrish 2003) and cyprinids (Scott & Crossman 1973). Our results, and those of Brown et al. (2002) suggest that the ability to use heterospecific alarm cues as ‘social information sources’ may be likewise impaired under weakly acidic conditions. Further experiments are needed to assess the potential impacts of the loss of heterospecific chemical cues. Magee et al. (2003) have demonstrated that both episodic and chronic exposures to acidified waters between pH 5.0 and 6.0 (intermediate range) results in significantly increased mortality in Atlantic salmon (Salmo salar) smolts as they migrate from fresh to salt water. They report no such increase in mortality in smolts exposed to weakly acidic to near neutral conditions (pH 6.0-6.6). This increase in smolt mortality appears to be due to physiological effect of intermediate acidification, which, while sublethal in freshwater, leads to increased mortality upon migration to salt water (Magee et al. 2003). Our results, (Brown et al. 2002; Leduc et al. 2003; Leduc et al., in press; this study), suggest that even in the absence of direct physiological effects, there may be additional indirect sublethal effects due to the loss of

Acquired predator recognition under weakly acidic conditions

111

function of conspecific and heterospecific chemical alarm cues. Experiments are ongoing to directly assess the fitness costs associated with this phenomenon.

Acknowledgements We would like to thank Drs. James Grant and Reehan Mirza for their comments on the manuscript, and Justin Golub, Mark Harvey, Patricia Foam and Isabelle Désormeaux for their help with the maintenance of fish. Financial support was provided by NSERC of Canada and Concordia University to G.E.B. All work reported herein was in accordance with Concordia University Animal Care Committee protocol #AC-2002-BROW.

References Atland A (1998) Behavioural responses of brown trout, Salmo trutta, juvenile in concentration gradients of pH and Al: a laboratory study. Env Biol Fish 53: 331–345 Baker JP, Bernard DP, Christensen SW, Sale MJ, Freda J, Heltcher K, Marmorek D, Rowe L, Scanlon P, Sutter G, Warren-Hicks W, Welbourn P (1990) Biological effects of changes in surface water acidic-base chemistry. National Acidic Precipitation Assessment Program, Acid Deposition: State of Science and Technology, Vol II NAPAP Rep. No. 13. USA, TN-Oak Ridge: U.S. Department of Energy, Oak Ridge Laboratory Barry KL, Grout AG, Levings CD, Nidle BH, Piercy E (2000) Impact of acid mine drainage on juvenile salmonids in an estuary near Britannia Beach in Howe Sound, British Columbia. Can J Fish Aquat Sci 57: 2332–2043 Brown C, Laland K (2001) Social learning and life skills training for hatchery reared fish. J Fish Biol 59: 471–493 Brown GE (2003) Learning about danger: chemical alarm cues and local risk assessment in prey fishes. Fish and Fisheries 4: 227–234 Brown GE, Godin J-GJ (1999) Who dares, learn: chemical inspection behaviour and acquired predator recognition in a characin fish. Anim Behav 57: 475–481 Brown GE, Smith RJF (1997) Conspecific skin extracts elicit antipredator responses in juvenile rainbow trout (Onchorhynchus mykiss). Can J Zool 75: 1916–1922 Brown GE, Smith RJF (1998) Acquired predator recognition in juvenile rainbow rout (Onchorhynchus mykiss): conditioning hatchery-reared fish to recognize chemical cues of a predator. Can J Fish Aquat Sci 55: 611–617 Brown GE, Adrian JC Jr., Patton T, Chivers DP (2001b) Fathead minnows learn to recognize predator odour when exposed to concentrations of artificial alarm pheromone below their behavioural-response threshold. Can J Zool 79: 2239–2245 Brown GE, Adrian JC Jr., Smyth E, Leet H, Brennan S (2000) Ostariophysan alarm pheromones: laboratory and field tests of the functional significance of nitrogen oxides. J Chem Ecol 26: 139–154 Brown GE, Adrian JC Jr., Kaufman IH, Erickson JL, Gershaneck D (2001a) Responses to nitrogen-oxides by Characiforme fishes suggest evolutionary conservation in Ostariophysan alarm pheromones. Pp 305–312 in Marchlewska-Koj A, Lepri JJ, Müller-Schwarze D (eds) Chemical Signals in Vertebrates, Vol. 9. USA, NY-New York: Plenum Press Brown GE, Adrian JC Jr., Lewis MG, Tower JM (2002) The effects of reduced pH on alarm signaling in Ostariophysan fishes. Can J Fish Aquat Sci 59: 1331–1338 Brown GE, Adrian JC Jr., Naderi NT, Harvey MC, Kelly JM (2003) Nitrogen oxides elicit antipredator responses in juvenile channel catfish, but not in convict cichlids or rainbow trout: conservation of the Ostariophysan alarm pheromone. J Chem Ecol 29: 1781–1796

Chemo-268.qxd

112

2/24/04 3:20 PM

Page 112

A.O.H.C. Leduc et al.

Chivers DP, Mirza RS (2001) Predator diet cues and the assessment of predation risk by aquatic vertebrates: a review and prospectus. Pp 277–284 in Marchlewska-Koj A, Lepri JJ, MüllerSchwarze D (eds) Chemical Signals in Vertebrates, Vol. 9. USA, NY-New York: Plenum Press Chivers DP, Smith RJF (1994) Fathead minnow, Pimephales premelas, acquired predator recognition when alarm substances is associated with the sight of unfamiliar fish. Anim Behav 48: 597–605 Chivers DP, Brown GE, Smith RJF (1998) Chemical alarm signaling in aquatic predator-prey systems: A review and prospectus. Ecoscience 5: 338–352 Chivers DP, Brown GE, Smith RJF (1996) The evolution of chemical alarm signals: Attracting predator benefits alarm signal senders. Am Nat 148: 649–659 Donnelly WA, Whoriskey FG (1993) Transplantation of Atlantic salmon (Salmo salar) and crypsis breakdown. Can Spec Publ Fish Aquat Sci 118: 25–34 Environment Canada (1997) Canadian Acid Rain Assessment Volume 3: The Effects on Canada’s Lakes, Rivers and Wetlands. CA-Quebec: Ministry of the Environment Gazdewich KJ, Chivers DP (2002) Acquired predator recognition by fathead minnows: influence of habitat characteristics on survival. J Chem Ecol 28: 439–445 Golub JL, Brown GE (2003) Are all signal the same? Ontogenetic changes in the response to conspecific and heterospecific chemical alarm signals by juvenile green sunfish (Lepomis cyanellus). Behav Ecol Sociobiol 54: 113–118 Haines TA, Baker JP (1986) Evidence of fish population responses to acidification in eastern United States. Water Air Soil Pollut 31: 605–629 Hesthagen T, Jonsson B (2002) Life history characteristics of brown trout in lakes at different stages of acidification. J Fish Biol 60: 415–416 Leduc AOHC, Kelly JM, Brown GE (2003) Detection of conspecific alarm cues by juvenile salmonids under neutral and weakly acidic conditions: laboratory and field tests. Oecologia (in press) Leduc AOHC, Noseworthy MK, Adrian JC,Jr, Brown GE (2003) Detection of conspecific and heterospecific alarm signals by juvenile pumpkinseed (Lepomis gibbosus) under weak acidic conditions. J Fish Biol 63: 1331–1336 Lorz HW, McPherson BP (1976) Effects of copper or zinc on the adaptation to sea water and ATPase activity and the effects of copper on migratory disposition of coho salmon (Onchorhyncus kisutch). J Fish Res Bd Can 33: 2023–2030

Received 15 September 2003; accepted 4 December 2003.

CHEMOECOLOGY Magee JD, Bedzinski M, McCormick SD, Kockick JF (2003) Effects of episodic acidification on Atlantic salmon (Salmo salar) smolts. Can J Fish Aquat Sci 60: 214–221 Magurran AE (1989) Acquired recognition of predator odour in the European minnows (Phoxinus phoxinus). Ethology 82: 216–233 Mathis A, Smith RJF (1993) Chemical alarm signals increase the survival time in fathead minnows (Pimephales promelas) during encounters with northern pike (Esox lucius). Behav Ecol 4: 260–265 Mirza RS, Chivers DP (2000) Predator-recognition training enhances survival of brook trout: evidence from laboratory and field-enclosure studies. Can J Zool 78: 2198–2208 Mirza RS, Chivers DP (2001) Are chemical alarm signals conserved within salmonids fishes? J Chem Ecol 27: 1641:1655 Mirza RS, Chivers DP (2003) Response of juvenile rainbow trout to varying concentrations of chemical alarm cues: response threshold and survival during encounters with predators. Can J Zool 81: 88–95 Raffenberg MJ, Parrish DL (2003) Interactions of Atlantic salmon (Salmo salar) and trout (Salvelinus fontinalis and Oncorhynchus mykiss) in Vermont tributaries of the Connecticut River. Can J Fish Aquat Sci 60: 279–285 Rodhe H, Langer J, Gallardo L, Kjellstorm E (1995) Global scale transport of acidifying pollutants. Water Air Soil Pollut 85: 37–50 Schindler DW, Mills KH, Malley DF, Findlay DL, Shearer JA, Davies OJ, Turner MA, Linsey GA, Cruikshank DR (1985) Long-term ecosystem stress: the effects of years of experimentally acidification on a small lake. Science 228: 1395–1401 Scott WB, Crossman EG (1973) Freshwater fishes of Canada. Fish Res Board Can Bull 184 Shively RS, Poe TP, Sauter ST (1996) Feeding response by northern squawfish to a hatchery release of juvenile salmonids in the Clearwater River, Idaho. Trans Am Fish Soc 125: 230–236 Smith RJF (1999) What good is smelly stuff in the skin? Cross function and cross taxa effects in fish ‘alarm substances’. Pp 475–488 in Johnston RE, Müller-Schwarze D, Sorensen PW (eds) Advances in Chemical Signals in Vertebrates. USA, NY-New York: Kluwer Academic Somers KM, Harvey HH (1984) Alteration of fish communities in lakes stressed by acid deposition and heavy metals near Wawa, Ontario. Can J Fish Aquat Sci 41: 20–29 Suboski MD, Templeton JJ (1989) Life skills training for hatchery fish: social learning and survival. Fish Res 7: 343–352