BREDER: FISH SCHOOLS AS OPERATIONAL STRUCTURES 



four fish lengths, both schools show a bulge on the 

 side closest to the other school and in so doing 

 automatically loosen their ranks slightly, but 

 sufficiently to allow the two bulges to merge, 

 forming a single school where there had been two. 

 This type of merging can usually be found be- 

 tween schools that do not differ in size by a factor 

 as large as four. 



The above describes what are evidently the 

 normal sequences to expect when two schools of 

 various size relationships have an encounter that 

 may lead to merging. This leads to the idea that 

 the "behavior" of a mirror image is sufficiently 

 unusual to prevent the further development of a 

 process leading to merging, the fishes evidently 

 recognizing a difference between another fish and 

 their own reflection. 



Sardinella anchovia Valenciennes and Brevoor- 

 tia patronus Goode avoided coming close enough 

 to the mirror for the development of any further 

 reaction. The first was present frequently in large 

 schools which tended to stay away from the dock 

 area in deeper water, but frequently came into the 

 shallower areas at which time they revealed no 

 indication of "nervousness." The second was seen 

 only as young fish in very small schools of not more 

 than 30 fast-moving individuals, that gave any 

 solid structure a wide berth, which is characteristic 

 of this species, at this place at least. Brevoortia in a 

 10-foot circular concrete tank formed a school of 

 about 30 individuals that averaged about 10 cm in 

 length. They had lived there for about 10 mo. 

 These fish were exposed to the mirror for 1 day in 

 August and 4 continuous days in November. Prior 

 to the introduction of the mirror, the school circled 

 the tank close to its wall. The introduction of the 

 mirror disrupted this path of the school which then 

 formed a tight mill as far away from the mirror as 

 possible. At no time were the fish observed to 

 approach the mirror. Only dropping food close to 

 its reflective surface would cause individuals to 

 move toward the mirror, and then only to snap at 

 the food and retreat rapidly. The fish fed less 

 during the presence of the mirror. After the 

 mirror was removed 6 days passed before the mill 

 broke up and the former swimming pattern was 

 resumed. Harengula pensacolae Goode and Bean, 

 not seen around the dock when the mirror was 

 used, behaved not unlike the Brevoortia in the 

 concrete pool. 



Mugil curema Valenciennes, in its very young 

 surface swarming stage of not over 2 cm, forms 

 very loose schools not at all like those of the adults. 



These young, on encountering the mirror, would 

 try persistently to swim into the mirror, seem- 

 ingly disregarding their mirror image that just as 

 persistently "opposed" them. Occasionally when 

 such a group left the mirror for reasons unknown, 

 a single fish would remain and continue to try to 

 swim through the mirror for a long period, 

 evidently almost to exhaustion. 



These observations were carried on from 8 June 

 to 10 September 1973, weather permitting, and 

 represent many repetitions of the facts and inter- 

 pretations. It is impossible to present these notes 

 in a more formal manner at this time. They clearly 

 have bearing on the present study and suggest the 

 desirability of going into this matter further as 

 another project which would in any case lead away 

 from present purposes. 



The observations indicate that there is a much 

 wider range of difference in response to the 

 mirror image than had been expected and there- 

 fore that the bonds that hold a school together are 

 not identical for each species, even if the total 

 result appears as a very similar geometric struc- 

 ture. It would seem that the response of a fish to a 

 fellow (here its mirror image) that approaches on a 

 true and unswerving collision course from which it 

 will not (cannot) budge is a truly frightening 

 experience. The difference in response between 

 Anchoa and Mugil in this case is especially strik- 

 ing. Anchoa acts in a manner that one might 

 anticipate, while the action of Mugil in placing 

 their mouths together has never been seen at any 

 age or size. 



LOCOMOTOR PROBLEMS 



With large numbers of fishes of one kind swim- 

 ming closely together in a common direction, the 

 locomotor needs of the participants would ob- 

 viously have influence on the structural nature of 

 the school, which in turn would also affect some 

 details of the locomotor efforts. Both classical and 

 contemporary hydrodynamics have to be invoked 

 in any attempt to understand this mechanical 

 aspect of school formation and operation. 



Flow Patterns 



To answer the question of whether water flow 

 induced by the propulsive activity of the fishes 

 themselves can help or hinder other fishes follow- 

 ing them depends on the direction and strength of 

 the flow and the angle of entry of a fish encoun- 



491 



