Interpopulation Variation in Mayfly Antipredator Tactics: Differential Effects of Contrasting Predatory Fish

Abstract

Introduced brown trout (Salmo trutta) have replaced native common river galaxia (Galaxias vulgaris) as the principal predators in many streams in the Taieri River system, New Zealand. Brown trout and common river galaxias present prey with contrasting predation risks as galaxias forage mainly at night using mechanical cues and trout present a higher predation risk during the day by foraging using visual cues. To determine if this change has affected the behavior of a siphlonurid mayfly, Nesameletus ornatus, we compared the behavior of N. ornatus nymphs from three neighboring streams that have different fish predation regimes (trout, galaxias, fishless) when confronted with all possible predation regimes in the laboratory (trout, galaxias, no fish). N. ornatus from a trout stream were consistently nocturanally active whether trout, galaxias, or no fish were present. In contrast, when mayflies from streams without trout were confronted with trout, no consistent diel periodicities in their behavior were observed. However, when mayflies from stream without trout were tested with and without galaxias they altered their behavior according to the presence/absence of the native predator. The possibility that N. ornatus accumulates evolutionary experience of trout through adults flying between streams may be ruled out: mayflies with no possible experience of trout did not respond differently to those from streams without trout but with the possibility of migrating from trout streams close by. The presence of different predatory fish may have resulted in differential trade—offs between foraging and predator avoidance. The probability of mayflies leaving food patches was higher during the day in the presence of trout. In contrast, the probability was higher during the night in the presence of galaxias. Trout and galaxias were also associated with differential reductions in the number of mayflies on the substrate surface and the proportion of mayflies on food patches, with consequent effects on time spent foraging. These responses indicate that both the history of the prey population and the nature of the predation risk have considerable influence on antipredator behavior and indicate a possible important influence on stream communities of predatory fish through sublethal effects on prey behavior.