Summary
Methodology
The study took place at the Queen's University Biological Field Station located on Lake Opinicon, Chaffey's Locks, Ontario. Fish were collected as eleutheroembryos (Balon, 1975) from nests of known parents in Lake Opinicon and brought immediately to the laboratory for rearing. The term 'fry' in this paper refers to a fish that is free-swimming and in its first summer of life. Fry were reared and observed in 96 1 wooden tanks with glass fronts. All tanks were supplied with a continuous flow of lake water, under a natural photoperiod. Water temperature was within ± 2°C of ambient lake water, and a 2 cm layer of gravel covered the bottom. Once fry were free-swimming, they were fed cultured live zooplankton three times per day.
Data collected on each species included a description of various agonistic MAPs, the frequency of occurrence of the MAPs, and the type of social organization which appeared. Once the agonistic behaviour appeared, individual fry were marked by branding the caudal fin (McNicol & Noakes, 1979). Largemouth bass fry were identified by pigmentation patterns on their caudal fin. Fry were measured (total length) once agonistic behaviour appeared. Four rock bass and largemouth fry and eight pumpkinseed and bluegill fry of similar size were maintained in the tanks, as higher densities resulted in an increase in mortality (pers. obs.).
The activities of the four species prior to the onset of the agonistic behaviour were determined. In rock bass, pumpkinseed, and bluegill fry, these activities consisted of feeding (Brown & Colgan, 1984) and aggregating, while in largemouth bass feeding and swarming were the major activities. Swarming was defined as a group of three or more fry oriented and swimming in the same direction and separated from one another by no more than 1 cm. Time spent swarming was recorded from the start of free-swimming for largemouth bass. The term 'swarming' was used because details of synchrony (turning, acceleration) were not recorded, and synchrony is an essential criterion for schooling (Pitcher, 1983).
In 1980, we described the development of the agonistic behaviour and social organizations. Daily observations were made on two tanks of each species from the onset of free-swimming until September 1. Five (four for largemouth and rock bass) individuals were observed for two minutes each, using the focal animal technique (Altmann, 1974), and the frequency of all MAPs was recorded. Dominance in encounters between individuals was based on the appearance of appeasement MAPs or fleeing. Individuals not displaying these MAPs in an encounter were considered dominant.
In 1981, we collected data on the total frequencies of the agonistic MAPs. Ten minutes of all occurrence sampling (Lehner, 1979) was carried out for each tank. Three tanks of largemouth fry, and two tanks of rock bass, pumpkinseed, and bluegill fry were observed daily. We recorded the relative sizes of individuals winning encounters for all agonistic encounters and the type of social organization which developed. In both years observations began one half to one hr after feeding. One of us (JAB) sat quietly 45 cm in front of a tank and began observations when fry returned to normal activities. Data were recorded on a prepared data sheet.
Daily observations were carried out in Lake Opinicon using SCUBA or snorkel. Approximately one hr per day was spent observing fry in the lake from May to September from 1979 through 1981. Largemouth bass fry were observed from their initial appearance until mid-September. In addition, six 60 m transects (parallel to shore) were made twice per week from June through September 1980. Data were collected on: areas in the littoral zone the fry inhabited; group size of fry; and agonistic interactions between individuals.