Abstract
Pioneering studies into the effects of elevated CO2 on the behaviour of reef fishes often tested high-CO2 reared fish using control water in the test arena. While subsequent studies using rearing treatment water (control or high CO2) in the test arena have confirmed the effects of high CO2 on a range of reef fish behaviours, a further investigation into the use of different test water in the experimental arena is warranted. Here, we used a fully factorial design to test the effect of rearing treatment water (control or high CO2) and experimental test water (control or high CO2) on antipredator responses of larval reef fishes. We tested antipredator behaviour in larval clownfish Amphiprionpercula and ambon damselfish Pomacentrus amboinensis, two species that have been used in previous high CO2 experiments. Specifically, we tested if: (1) using control or high CO2 water in a two channel flume influenced the response of larval clownfish to predator odour; and (2) using control or high CO2 water in the test arena influenced the escape response of larval damselfish to a startle stimulus. Finally, (3) because the effects of high CO2 on fish behaviour appear to be caused by altered function of the GABA-A neurotransmitter we tested if antipredator behaviours were restored in clownfish treated with a GABA antagonist (gabazine) in high CO2 water. Larval clownfish reared from hatching in control water (496 mu atm) strongly avoided predator cue whereas larval clownfish reared from hatching in high CO2 (1,022 m atm) were attracted to the predator cue, as has been reported in previous studies. There was no effect on fish responses of using either control or high CO2 water in the flume. Larval damselfish reared for four days in high CO2 (1,051 mu atm) exhibited a slower response to a startle stimulus and slower escape speed compared with fish reared in control conditions (464 m atm). There was no effect of test water on escape responses. Treatment of high-CO2 reared clownfish with 4 mg l(-1) gabazine in high CO2 seawater restored the normal response to predator odour, as has been previously reported with fish tested in control water. Our results show that using control water in the experimental trials did not influence the results of previous studies on antipredator behaviour of reef fishes and also supports the results of novel experiments conducted in natural reef habitat at ambient CO2 levels.