Search

Summary 1. Field observations indicate that the ability to feed at different light intensities may differ between brown trout and Arctic charr, and this is the first study to test this experimentally. To establish a background level of feeding in daylight at midday, trout and charr in two size groups were kept in tanks (one fish per tank) at three constant temperatures (5.0, 10.8 and 13.0 °C) and each fish was offered, one at a time, 50 freshly killed shrimps ( Gammarus pulex), the number eaten being recorded. Shrimps could only be taken in the water column because a metal mesh prevented access to dead shrimps on the tank bottom. In a first series of experiments, individual fish were kept at one of 10 natural light intensities (range 0.001-50 lx). In a second series, conditions were similar except that the fish tank was covered in black polyethylene and had a light-tight lid with a shutter so that light levels could be kept constant, using artificial illumination. In a third series, the fish were fed in total darkness, but the false bottom was removed, allowing access to dead shrimps on the tank bottom as well as in the water column. 2. The results of the first and second series differed interspecifically but were very similar intraspecifically, with no significant differences between the food intake for the two size groups or in the experiments at 10.8 and 13.0 °C. Food intake remained fairly constant at light intensities between 50 lx (dusk or dawn) and 0.03 lx and was similar to that of fish feeding at midday. At 10.8 and 13.0 °C, food intake between 0.03 and 50 lx was higher for trout than for charr, mean values for shrimps eaten per fish being 39.9 for trout (range 36-44, n = 100 fish) and 32.0 for charr (range 28-38, n = 100), but at 5.0 °C, the situation was reversed with mean values of 15.1 for trout (range 11-18, n = 50 fish) and 19.8 for charr (range 17-22, n = 50). 3. As light intensity decreased from 0.04 to 0.001 lx, feeding rate decreased exponentially but was always higher for charr than for trout, with a mean number of shrimps eaten at 0.001 lx of 9.3 for trout (range 5-13, n = 20 fish) and 13.6 for charr (range 9-20, n = 20) at 10.8 and 13.0 °C, and 2.0 for trout (range 1-4, n = 10 fish) and 5.5 for charr (range 2-8, n = 10) at 5.0 °C. In total darkness (false bottom fitted), none of the 50 shrimps was taken by either species. When the false bottom was removed in the third series, the mean number of shrimps consumed over 24 h was eight for trout (range 3-11, n = 20 fish) and 14.9 for charr (range 9-20, n = 20) at 10.8 and 13.0 °C, and two for trout (range 0-4, n = 10 fish) and five for charr (range 3-8, n = 10) at 5.0 °C. 4. Therefore, the feeding ability of trout was superior to that of charr when using photopic vision in daylight and mesotopic vision at dusk and dawn, but inferior to that of charr when using scotopic vision at low light intensity. Charr were also superior at low temperatures and when foraging for food in total darkness. Therefore, as light intensity decreases after dusk in their natural habitat, the advantage in feeding will shift from trout to charr, with the reverse occurring as light intensity increases after dawn. [ABSTRACT FROM AUTHOR]