Premium
Danger at dawn: experimental support for the twilight hypothesis in shoaling minnows
Author(s) -
Pitcher T. J.,
Turner J. R.
Publication year - 1986
Publication title -
journal of fish biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.672
H-Index - 115
eISSN - 1095-8649
pISSN - 0022-1112
DOI - 10.1111/j.1095-8649.1986.tb04999.x
Subject(s) - shoaling and schooling , dusk , twilight , predation , daytime , biology , fishery , fish <actinopterygii> , moonlight , nocturnal , predator , pike , foraging , sunset , ecology , atmospheric sciences , physics , astronomy , geology
This experiment aimed to compare responses to predator threat at dawn and in daylight. Using silicon‐intensified video, shoaling European minnows, Phoxinus phoxinus (L.), were observed in large tanks furnished to simulate natural conditions and equipped with artificial dawn and dusk illumination. In the dark, fish were observed under infra‐red light. The appropriate light level for dawn predator exposure was determined behaviourally. During dawn, the behaviour of individual focal minnows was systematically recorded at five light levels set by the behaviour of fish in pilot work. On six predetermined dawn and daytime occasions over a period of 3 weeks, minnows feeding on an artificial vertical feeding patch were subjected to a standardized stalk by a model pike. Under daytime illumination (8 μEm −2 s −1 ), the minnows were presented with patches once per day. The feeding patches were presented early in the artificial dawn (0.5–1.5 μm −2 s −1 ) only on days when fish were to be subjected to the model pike. Fish became active at the onset of dawn. A behaviour which appeared at this time is reported for the first time: minnows rose to the surface, appeared to take in and expel bubbles of air, and descended. Shoaling responses and swimming speeds increased soon after the start of dawn. Foraging increased slightly later than this, when illumination had reached about one‐fifth of the daytime level. This was the level chosen for predator exposure. When approached by the pike model at dawn, minnows abandoned feeding on the patch only when the pike reached one‐third of the distance which it obtained under daylight. Individual fish detecting the approaching threat showed skitter and inspection behaviour similar to that observed in daylight, but these behaviours peaked much later in the pike's stalk than in daylight. We conclude that minnows at dawn fail to detect the approaching pike until it is much closer than in daylight; we believe that the work therefore provides the first experimental support for the ‘twilight hypothesis’ that, under low light levels, stalking predators have an advantage over shoaling prey.