Sculpin numbers increased dramatically over the course of
the salmon spawning run on Woody Island and decreased in a
less dramatic fashion after spawning completion (Fig. 1A).
The increase was noted days before the commencement of
sockeye spawning and was apparent only on the spawning
beach itself and not on directly adjacent habitats. This spawning
site specific increase suggests that sculpins may predict the
arrival of the egg resource, and do not simply react to its presence.
We know from in situ experiments on Woody Island that
sculpins are highly attracted to the odor of fresh sockeye eggs
(Dittman et al. 1998) and to the digging actions of females
during nest construction (C.J. Foote and G.S. Brown, personal
observation; see also Janssen 1990). However, neither event
occurred during the first 6 days of our observations, yet sculpin
numbers still increased. Salmon increasingly gathered and
schooled around the spawning area before settling, but their
movements were broad and carried them all around the inner
bay, including nonspawning areas. The fact that sculpin numbers
increased only in the eventual sockeye spawning area
suggests that visual cues of the positions of the schooling sockeye
were also not the cues for directed movement by the
sculpins. However, it is possible that sculpins were detecting
residual cues of previous spawning in the area, as has been
shown in lake trout (Salvelinus namaycush) (Foster 1985), or
had simply learned where the spawning area was from previous
experience. Sculpins live up to 8 years in Iliamna Lake
(Roger 1971), and hence have repeated opportunities to learn
the temporal and spatial distribution of egg resources.