Summary
Studies on the ontogeny of behaviour in fish have seldom considered the adaptive significance of the order of appearance of the behaviours. Results of laboratory studies and field observations on the ontogeny of feeding, predator-avoidance, and agonistic behaviour in largemouth bass, Micropterus salmoides, rock bass, Ambloplites rupestris, and pumpkinseed, Lepomis gibbosus, young indicate that the order of appearance of the various behavioural systems enhances the survival of the young. In the laboratory, all species spend significantly more time feeding during the first weeks of free-swimming than the later weeks. During a corresponding period in the field the young are either occupying an offshore area low in predators (rock bass, pumpkinseed) or are being guarded by a parent (largemouth bass); thus the risk of predation is low. When the young are in a predator-rich environment (inshore) both the predator-avoidance response and agonistic behaviour are well developed. Agonistic behaviour is the last to appear and may serve to disperse the young. Dispersal may relate to the feeding mode of the various species and may also reduce the probability of predation. It is apparent that the sequential onsets of the behavioural systems are in concert with ecological events and selective pressures confronting centarchid young.
Methodology
The study was conducted at the Queen's University Biological Field Station located on Lake Opinicon, Chaffey's Locks, Ontario, Canada. Fish were collected as eleutheroembryos (Balon 1975) from nests and immediately placed in separate wooden 90 1 glass-fronted tanks with a 2 cm layer of gravel on the bottom. All tanks were supplied with a continuous flow of lake water and a natural photoperiod was maintained. Water temperature was within 2° C of ambient.
Free-swimming young were fed culture live zooplankton three times per day (see Brown & Colgan 1984). Plankton was collected from the lake and placed in two 3 m diameter x 36cm deep plastic pools filled with lake water. The pools were fertilized to enhance growth of phytoplankton. Plankton was collected and added to pools weekly. The pool water was monitored every other week in May and June and weekly in July and August, to determine the type (family) and approximate density of plankton. In addition, the constant flow of lake water to each tank contained food items. Preliminary studies had indicated that largemouth bass young grew more slowly and suffered higher mortality past seven weeks of age if fed just plankton. Thus the diet of the largemouth bass young was supplemented with frozen brine shrimp, Artemia salina, from eight weeks onwards and only the first seven weeks of foraging time were recorded for largemouth bass young. Densities of the young in the tanks varied from 150 individuals initially and were reduced to 30 per tank by the end of the study period. The term young refers to fish that have commenced free-swimming and are less than one year old, i.e. encompasses larvae and first year juveniles (Balon 1975).