The fast‐start escape response in fish is essential for predator avoidance, but almost nothing is known about whether sublethal concentrations of pollutants can impair this reflex. Ammonia, a pervasive pollutant of aquatic habitats, is known to have toxic effects on nervous and muscle function in teleost fish. Golden gray mullet (Liza aurata L.) were exposed for 24 h to sublethal ammonia concentrations in seawater (control, 400 μmol L−1, or 1,600 μmol L−1 NH4Cl), and then their response to startling with a mechanical stimulus was measured with high‐speed video. Initiation of the escape response was significantly slowed by ammonia exposure: response latency rose proportionally from <50 ms in controls to >300 ms at a concentration of 1,600 μmol L−1 NH4Cl. This indicates toxic effects on nervous function within the reflex arc. Impaired escape performance was also observed: maximum turning rate, distance covered, velocity, and acceleration were significantly reduced by >45% at a concentration of 1,600 μmol L−1 NH4Cl. This indicates toxic effects on fast‐twitch glycolytic white muscle function, the muscle type that powers the fast‐start response. These neuromotor impairments were associated with significant ammonia accumulations in venous plasma and white muscle and brain tissue. These results indicate that anthropogenic ammonia pollution in aquatic habitats may increase the vulnerability of fish to predation, especially by birds and mammals that are not affected by water ammonia concentrations.

Hypoxia is a phenomenon occurring in marine coastal areas with increasing frequency. While hypoxia has been documented to affect fish activity and metabolism, recent evidence shows that hypoxia can also have a detrimental effect on various antipredator behaviours. Here, we review such evidence with a focus on the effect of hypoxia on fish escape responses, its modulation by aquatic surface respiration (ASR) and schooling behaviour. The main effect of hypoxia on escape behaviour was found in responsiveness and directionality. Locomotor performance in escapes was expected to be relatively independent of hypoxia, since escape responses are fuelled anaerobically. However, hypoxia decreased locomotor performance in some species (Mugilidae) although only in the absence of ASR in severe hypoxia. ASR allows fish to show higher escape performance than fish staying in the water column where hypoxia occurs. This situation provides a trade-off whereby fish may perform ASR in order to avoid the detrimental effects of hypoxia, although they would be subjected to higher exposure to aerial predation. As a result of this trade-off, fishes appear to minimize surfacing behaviour in the presence of aerial predators and to surface near shelters, where possible.For many fish species, schooling can be an effective antipredator behaviour. Severe hypoxia may lead to the disruption of the school unit. At moderate levels, hypoxia can increase school volume and can change the shuffling behaviour of individuals. By altering school structure and dynamics, hypoxia may affect the well functioning of schooling in terms of synchronization and execution of antipredator manoeuvres. School structure and volume appear to be the results of numerous trade-offs, where school shape may be dictated by the presence of predators, the need for energy saving via hydrodynamic advantages and oxygen level.The effects of hypoxia on aquatic organisms can be taxon specific. While hypoxia may not necessarily increase the vulnerability of fish subject to predation by other fish ( since feeding in fish also decreases in hypoxia), predators from other taxa such as birds, jellyfish or aquatic mammals may take advantage of the detrimental effects of hypoxia on fish escape ability. Therefore, the effect of hypoxia on fish antipredator behaviours may have major consequences for the composition of aquatic communities.

SUMMARY We examined the effect of turbidity (0.5–14 beam attenuation m–1) and predator attack speed (150 and 296 cm s–1) on escape responses of juvenile cod Gadus morhua in the laboratory. We triggered escape responses using a predator model and measured escape timing, direction and locomotor performance. We also measured responsiveness and estimated the likelihood of fish escaping the `predator attack' (putative escape success, PES). Turbidity affected both PES and the type of escape response used by the fish, but these effects depended on predator speed. PES for the fast predator attack declined from 73% in clear water to 21% in highly turbid water, due to decreased responsiveness and poorly timed escapes. Intermediate turbidity enhanced PES and responsiveness to the slow predator attack. Locomotor performance was reduced by turbidity, whereas predator speed had the opposite effect. Our results suggest that both predator attack speed and turbidity have important roles in determining the vulnerability of fish attacked by piscivorous predators.

In hypoxia, gray mullet surface to ventilate well-oxygenated water in contact with air, an adaptive response known as aquatic surface respiration (ASR). Reflex control of ASR and its behavioral modulation by perceived threat of aerial predation and turbid water were studied on mullet in a partly sheltered aquarium with free surface access. Injections of sodium cyanide (NaCN) into either the bloodstream ( internal) or ventilatory water stream ( external) revealed that ASR, hypoxic bradycardia, and branchial hyperventilation were stimulated by chemoreceptors sensitive to both systemic and water O-2 levels. Sight of a model avian predator elicited bradycardia and hypoventilation, a fear response that inhibited reflex hyperventilation following external NaCN. The time lag to initiation of ASR following NaCN increased, but response intensity ( number of events, time at the surface) was unchanged. Mullet, however, modified their behavior to surface under shelter or near the aquarium edges. Turbid water abolished the fear response and effects of the predator on gill ventilation and timing of ASR following external NaCN, presumably because of reduced visibility. However, in turbidity, mullet consistently performed ASR under shelter or near the aquarium edges. These adaptive modulations of ASR behavior would allow mullet to retain advantages of the chemoreflex when threatened by avian predators or when unable to perceive potential threats in turbidity.

Escape performance was investigated in the golden grey mullet Liza aurata exposed to various levels of oxygen: >85 (i.e. normoxia), 50, 20 and 10 % air saturation. Since the golden grey mullet performed aquatic surface respiration when air saturation approached 15–10 %, escape performance was tested at 10 % air saturation with and without access to the surface (10 % S and 10 % C, respectively). Various locomotor and behavioural variables were measured, such as cumulative distance, maximum swimming speed, acceleration, responsiveness (per cent of responding fish), response latency and directionality. Golden grey mullet showed a decrease in responsiveness when the oxygen level was reduced to 10 % air saturation, whether the surface access was obstructed or not. Hypoxia did not have any effect on the response latency. Cumulative distance and maximum swimming speed over a fixed time were significantly different between normoxic conditions and 10 % C, while no differences were found in maximum acceleration. While the fish's ‘C‐bend’ was mainly directed away from the stimulus in normoxia, the proportion of away and towards ‘C‐bend’ was random when the oxygen was ≤20 % air saturation. This suggested an impairment of the left‐right discrimination at the initiation of the fast start. Hypoxia affected golden grey mullet escape performance mainly through an impairment of responsiveness and directionality, while locomotor performance was affected only in severe hypoxia when the surface was obstructed. The study showed that, in addition to forcing the fish to the surface as shown by previous studies, hypoxia may also reduce fish elusiveness facing a predator by directly impairing its escape performance.

The proposal that plasma ammonia accumulation might impair the swimming performance of fish was first made over a decade ago, and has now proven to be the case for a number of salmonid species. The first experimental evidence was indirect, when a negative linear relationship between plasma ammonia concentrations and maximum sustainable swimming speed (U(crit)) was found following the exposure of brown trout (Salmo trutta) to sub-lethal concentrations of copper in soft acidic water. Since then, negative linear relationships between plasma ammonia concentration and U(crit) have been demonstrated following exposure of brown trout, rainbow trout (Oncorhynchus mykiss) and coho salmon (Oncorhynchus kisutch) to elevated water ammonia. For brown trout, the relationships between plasma ammonia and U(crit) were remarkably similar following either exposure to elevated water ammonia or to sub-lethal copper. This indicates that the impairment of swimming performance resulting from exposure to sub-lethal concentrations of heavy metals may be attributable in large part to an accumulation of endogenous ammonia. The negative relationship between plasma ammonia concentration and U(crit) was similar in size-matched rainbow and brown trout but, under similar regimes of ammonia exposure, rainbow trout were able to maintain a significantly lower plasma ammonia concentration, revealing inter-specific differences in ammonia permeability and/or transport. One primary mechanism by which ammonia accumulation may impair exercise performance is a partial depolarisation of membrane potential in tissues such as the brain and white muscle. This may prejudice the co-ordination of swimming movements and reduce or abolish the development of muscle tension, thus, compromising swimming efficiency and performance at the top end of the range.

Adult trout Oncorhynchus mykiss fitted with a dorsal aortic catheter were exposed to 288+/-15 micromol l(-1) (mean +/- S.E.M.) total ammonia for 24h in water at a pH of 8.39+/-0.02, while swimming at a speed equivalent to 0.75 bodylengths s(-1) (BLs(-1)) in a Brett-type tunnel respirometer. The fish were then exposed to stepwise increments in swimming speed (0.25 BLs(-1) every 30 min) until exhaustion. Measurements of oxygen uptake (M(O2)) and plasma total ammonia levels and pH were made at each speed. Control trout were treated identically but without exposure to ammonia. Ammonia exposure caused an increase in plasma total ammonia level to 436+/-34 micromol l(-1), compared to 183+/-30 micromol l(-1)in control animals (N=6). A significant reduction in total plasma ammonia level was found in both groups during exercise, despite a large negative concentration gradient in those exposed to an elevated concentration of ammonia in water, which may indicate an active excretory process. The overall increase in plasma ammonia levels in exposed trout was associated with a significant reduction in critical swimming speed (U(crit)) to 1.61+/-0.17BL s(-1) from 2.23+/-0.15BL s(-1) in control animals. Ammonia-exposed trout had a significantly higher maintenance metabolic rate (MMR) than control fish, when estimated as the y-intercept of the relationship between swimming speed and M(O2). Active metabolic rate (AMR, maximum M(O2) as measured at U(crit)) was significantly lower in ammonia-exposed animals, leading to a profound reduction in factorial aerobic scope (AMR/MMR). Reduced U(crit) was also linked to a reduction in maximum tailbeat frequency. Calculation of membrane potentials (E(M)) in the white muscle of fish swum to U(crit) revealed a significant partial depolarisation of white muscle in ammonia-exposed fish. This may have prevented white muscle recruitment and contributed to the reduced maximum tailbeat frequency and overall impairment of swimming performance in the ammonia-exposed fish.