sented three levels of light intensity (high, 

 medium, and low) and two light positions (above 

 and below the surface). Each combination of 

 lighting characteristics was replicated four 

 times — two with the light on the left during the 

 18-hour interim and two with the light on the 

 right. Table 5 gives the scores of the trials, before 

 and after the 18-hour interim. The number of fish 

 in the lighted zone was reduced significantly after 

 18 hours (F = 13.4, P<0.01). This reduction was 

 significantly greater when the lights were above 

 the surface (F= 11.83, P<0.01) than when the 

 lights were below, but the differences in re- 

 duction associated with light intensity were not 

 significant. 



T,\BLE 5. — Comparison of scores  for light attraction before 

 and after an 18-hour interval of constant stimulus, in 

 relation to light location and intensity (experiment 9) 



Before interval 



After interval 



Light intensity 



Light 

 above 

 surface 



Light 

 below 

 surface 



Mean 



Light 

 above 

 surface 



Light 

 below 

 surface 



Mean 



High 109 



144 



109 

 158 



Medium 



Low 



Mean 143 



138 

 132 

 133 



149 



134 



94 

 80 

 57 

 126 



130 

 125 

 114 

 147 



' Score = sum of numbers of herring counted in the illuminated half of the 

 tank at 30 1-minute intervals. Ten herring were used in each test; maximum 

 possible score = 300. 



EFFECT OF OXYGEN CONCENTRATION 



The possible imix)rtance of oxygen concentra- 

 tion to the light i-esponse was suggested partly by 

 observations that the habitat of juvenile heiTing 

 was frequently supersaturated with oxygen in sum- 

 mer (Colton, Marak, Nickerson, and Stoddard, 

 1968; Stickney, 1968) and partly by a comment of 

 Kalle ( 1965) that the vertical migration of herring 

 (usually considered a resix)nse to light) might be 

 due to depletion of oxygen in dense schools near 

 the bottom. 



Experiment 10. Groups of 10 herring were tested 

 on alternate days in water normally saturated or 

 highly supersaturated with oxygen before each 

 trial. The temperature of the experiment was 15.5 

 to 17° C. ; the mean oxygen levels were 8.1 p.p.m. 



(98 percent saturation) and 17.2 p.p.m. (212 per- 

 cent saturation). The herring were taken from 

 storage at 11 to 14° C, 110 to 125 percent 0, satura- 

 tion. Each trial consisted of the routine exposure 

 to an attracting light; scoi-es were based on the 

 number of fish counted each minute for 30 minutes 

 in the lighted end of the tank. The mean score for 

 six trials in normal water was 122; that for six 

 trials in suiiereaturated water was 164. The dif- 

 ference had a "t" value of 2.04 (P<0.1) ; treated 

 as paired data, the differences between each pair 

 of trial scores had a "t" value of 4.28 (P<0.01). 

 The experiment seemed to indicate that supersatu- 

 ration had a significant effect on the attraction of 

 the herring to light. Subsequent experiments (11 

 to 13) did not cori-oborate these results, however. 



Experiment 11. Herring were taken from hold- 

 ing tanks at 10 to 14° C. and 105 to 130 percent 

 saturation of oxygen and held 6 hours under the 

 experimental conditions before each trial. Nor- 

 mally saturated and supersaturated water averag- 

 ing 9.6 p.p.m. (93 percent saturation) and 19.7 

 p.p.m. (185 percent saturation) of oxygen, respec- 

 tively, at 6 to 9° C. were, provided on alternate 

 days. In five triaJs at each oxygen level, mean 

 scores were 198 in the normal water, 205 in the 

 supersaturated water. The difference is not sig- 

 nificant (t=0.43). 



Exjjeriment 12. This experiment was actually a 

 part, of experiment 15 and included the additional 

 variable of light-dark adaptation. The three levels 

 of oxygen concentration were low (mean = 5.3 

 p.p.m., 63 percent saturation), saturated (mean 

 = 7.9 p.p.m., 93 percent saturation), and super- 

 saturated (mean = 21.8 p.p.m., 240 percent satura- 

 tion). Herring which had been adapted 6 hours to 

 light or darkness before each trial were tested at 

 each level of oxygen, making six combinations of 

 experimental variables, each combination repli- 

 cated five times. The herring were taken from stor- 

 age at 12 to 17° C. and 100 to 124 percent 

 saturation of oxygen and exposed 6 hours to the 

 experimental conditions before each trial. The 

 mean scores for ten trials at each level of oxygen 

 were 125, 138, and 123. The difference among them 

 was not significant (F=0.7). 



Experiment 13. The effects of light or dark 

 adaptation were most pronounced during the first 

 minute of exposure to the attracting light. Because 

 this critical time period may not have been ade- 



80 



U.S. FISH AND WILDLIFE SERVICE 



