The effects of temperature on growth, pelagic larval duration (PLD) and maximum swimming speed were compared in the tropical fish marine species Amphiprion melanopus, to determine how temperature change affects these three factors critical to survival in larvae. The effects of rearing temperature (25 and 28 degreesC) on the length of the larval period and growth were examined in conjunction with the effects of swimming temperature (reared at 25 degreesC, swum at 25 and 28 degreesC, reared at 28 degreesC, swum at 25 and 28 degreesC) on critical swimming speed (U-crit). Larvae reared at 25 degreesC had a 25% longer pelagic larval duration (PLD) than larvae reared at 28 degreesC, 12.3 (+/- 0.3) days compared with 9 (+/- 0.6) days at 25 degreesC. To offset this effect of reduced developmental rate, growth and U-crit were measured in larvae reared at 28 and 25 degreesC at the same absolute age (7 days after hatching (dah)) and same developmental age (7 dah at 28 degreesC cf. 11 dah at 25 degreesC), corresponding to the day before metamorphosis. Larvae reared at 25 degreesC were smaller than larvae reared at 28 degreesC at the same absolute age (7 dah at 25 degreesC cf. 7 dah at 28 degreesC), yet larger at similar developmental age (11 dah at 25 degreesC cf. 7 dah at 28 degreesC) when weight and standard length were compared. This stage-specific size increase did not result in better performance in larvae at the same developmental age, as there was no difference in U-crit in premetamorphic larvae reared at either temperature (7 dah at 28 degreesC c.f 11 dah at 25 degreesC). However, U-crit was considerably slower in 7-day-old larvae reared at 25 degreesC than larvae of the same absolute age (7 dah) reared at 28 degreesC. Swimming temperature controls demonstrated that a change in temperature immediately prior to swimming tests did not effect swimming performance for larvae reared at either temperature. A decreased in rearing temperature resulted in longer larval durations, reduced growth rates and slower swimming development in larvae. However, the magnitude of the response of each of these traits varied considerably. As such, larvae reared at the lower temperature were a larger size at metamorphosis but had poorer relative swimming capabilities. This study highlights the importance of measuring a range of ecologically relevant traits in developing larvae to properly characterise their relative condition and performance in response to environmental change. (C) 2003 Elsevier B.V. All rights reserved.
