ON THE COMPETITION BETWEEN WHITEFISH SPECIES 335 



COMPETITION IN THE ADULTS 



Turning now to the adult fish, the design of their ecological isolation seems 

 much easier to reveal than that of the younger stages. This is in a way- 

 puzzling as the most effective segregating mechanisms should be expected 

 during the time they are most wanted, namely in the early stages when the 

 size of the populations are mainly decided. In contrast, the adults display 

 obvious specific food preferences that are recombined and modified from 

 lake to lake (Figs. 6 and 7). This phenomenon is made possible by a food 

 choice mechanism. This mechanism involves a temporal learning to concen- 

 trate on specific food objects as they give the best reward for the energy 

 spent (cf Allen, 1941; Nilsson, 1955, 1957 and 1958; Hinde, 1959), and first 

 seems to become fully effective when the fish has reached a certain age or size. 



C. pidschian displays the most fixed pattern in feeding habits; it has in 

 both cases an intermediate growth (Fig. 8) and is mainly a bottom feeder 

 with strong exploitation of chironomid larvae, caddis fly larvae, Pisidium, 

 Limnaea and bottom-dwelling crustaceans. C. peled, on the other hand, 

 appears in Vojmsjon as a dwarfed plankton-eater (with a diet also comprising 

 some insect-food), in Storavan-Uddjaur as a fast-growing insect-eater. 

 C. lavaretus, which is not present in Vojmsjon has in Storavan-Uddjaur on 

 the whole taken over the food niche occupied by C. peled in Vojmsjon and 

 appears as a plankton-eater of ciscoe type. 



When trying to interpret these facts it should be observed that food 

 competition is difficult to trace in fish, as feeding habits of fish are mostly 

 very flexible and the food niches of different species very often considerably 

 overlapping. To measure the degree of competition by measuring the degree 

 of similarity in feeding habits between sympatric fish species must be wrong 

 as similarity is often rather a sign of superabundance of food (cf. Nilsson, 

 1955)* Noi^ is it of course correct indiscriminately to interpret dissimilarity 

 as a sign of competition as it might as well often be a sign of what Brian 

 (1956) calls selective segregation, i.e. a segregation most frequently occurring 

 between phylogenetically distant forms, that fundamentally should each 

 keep to the same habitats or feeding habits whether living sympatrically or 

 not, and whether being restricted to limited supplies or not. A better method 

 might be to compare feeding habits as well as other ecological characteristics 

 of closely related species when hving allopatrically and sympatrically in 

 different combinations. Moreover it is necessary to study the variations in 

 feeding habits between years, seasons and hfe stages (cf Figs. 6, 7 and 9). 



The present authors, in agreement with Svardson (1954) and in analogy 

 with the fmdings from other food selection studies (cf e.g. Lack, 1947; 

 Gibb, 1954; Nilsson, 1955) interpret the differences in feeding habits between 



