to include those circumstances favorable to light 

 attraction and those unfavorable. 



The difference in resjDonse to light shown by 

 individual fish and by fish in groups under other- 

 wise similar conditions may be another significant 

 aspect of behavior. The data from experiment 1, 

 in which the responses of 90 individual herring to 

 light were tested at three different temperatures, 

 and from experiment 6, in which the responses of 

 90 herring were tested in groups of 10 at the same 

 three temperatures are an example. If the fish 

 tested individually are combined arbitrarily into 

 three gi-oups of 10 for each temperature, and the 

 mean score for each group is expressed as a percent- 

 age of the maximum possible score, measures of 

 variance among the groups can be calculated. Simi- 

 larly, the variance among the scores, expressed as a 

 percentage of the maximum possible score, can be 

 calculated for the actual groups of 10 fish observed 

 in experiment 6. A comparison shows that the 

 variance among the scores of the actual groups of 

 10 fish is significantly greater than the variance 

 among the scores of the arbitrarily created groups 

 of fish tested individually. In fact, the variance 

 among the scores of the actual groups is not sig- 

 nificantly different from the variance among the 

 scores of individuals. These facts imply that the 

 collective response of 10 herring in a group is not 

 simply an average of their uidividual responses. 

 Instead, the collective response seems to reflect the 

 individual responses of only one or two fish in the 

 group. 



To explain the apparent lack of thermal in- 

 fluence on fisl^ individually in contrast to the 

 significant thennal influence on groups, I suggest 

 the following hypothesis. Most herring are only 

 feebly influenced by temperature in their response 

 to light. The preponderance of fish in this category 

 causes the average response to appear uninfluenced 

 by temperature when each fish responds as an indi- 

 vidual, even though a few individuals may be 

 strongly influenced. When the fish are in groups, 

 however, the weakly influenced majority respond 

 not so much to the stimulus itself as to the strongly 

 influenced minority, whose behavior dominates 

 the group. I believe that in this interaction lies 

 II the significance of the school in fish behavior: 

 i| the interaction provides to the group a sensitivity 

 'I and an ability not possessed by individuals to 

 react in an unequivocal manner to a situation. 



FAOTORS INFLUENCING ATTRACTION OF ATLANTTIC HERRING TO AR(TIFICIAL LIGHTS 



RESPONSES OF HERRING TO LIGHT 

 AND THEIR APPLICATION IN THE 

 FISHERY 



Without doubt attraction to artificial lights at 

 night is a significant behavioral response of 

 herring, and it is potentially useful in the herring 

 fishery. The question is: Under what conditions 

 is this response brought out most strongly and 

 what tactics in using lights can be most effectively 

 employed ? 



Most of the evidence indicates that a sub- 

 merged light is more effective than one above the 

 surface. One reason is that the entire output of 

 the underwater light is used, whereas a large por- 

 tion of the light from above the water is reflected 

 from the surface. Tlie submerged light is also 

 more uniform : the rays do not flicker from the 

 effect of a ruffled surface. Moreover, the sub- 

 merged light has an improved attracting effect 

 which is independent of its gi-eater efficiency. A 

 submerged light which produced only 1/10 to 

 1/1,000 the underwater illumination of a light 

 above an unruffled surface proved to be the more 

 effective in laboratory experiments. Because of 

 refraction, the rays from a light above the surface 

 project sharply downward, even at some distance 

 from the light source. It may be that the direction 

 of the rays in relation to the position of the fish 

 are important, and that rays from above the sur- 

 face approaching the vertical tend to repel 

 herring. The light from the sun, sky, or a bright 

 moon would be of this nature; all of these light 

 sources tend to keep the herring from the surface 

 and may be the cause of the characteristic diurnal 

 vertical migrations of herring. 



E\ddence from my experiments and also from 

 other studies shows that the brightest lights are 

 not necessarily the most effective for attracting 

 herring. Although a brighter light will have a 

 greater range and can be seen by fish at a greater 

 distance, the illumination within a certain radius 

 may exceed the optimum and tend to repel the 

 fish even if they are attracted up to that radius. 

 To obtain maximum range while still attracting 

 nearby fish, certain manipulations of the light 

 have been used effectively. The simplest method 

 is to dim the light gradually (Gauthier, in press; 

 Kurc, in press; Strom, in press). Another scheme 

 was described by Sasaki (1959) : A series of lights 

 of optimum brilliance extend some distance from 



83 



