used for decoration) were also employedo In all experiments^ 

 the bulbs were hung on the tank side of baffle D (figo Ih) and 

 were practically or completely submerged o 



Preliminary studies showed that during daylight hours ohe 

 artificial light^, in contrast to sunlight,^ was too weak to 

 elicit definite response from the fisho Furthermorej even at 

 late afternoons the reactions were irregular because of various 

 distiirbances about the tanko Thereforej all the experiments 

 were performed after dusk or at night and at times when the 

 tide was sufficiently low t- insure clear water in the tank for 

 conparable light penetration. 



As the tuna fed regularly from the western side of the 

 middle of the southern half of the tank (at F in figo lii)s all 

 light stimuli were applied from the northern edge of the tank 

 and only when a tuna entered into a specifically selected area 

 or field in the northern one-quarter of the tank would its 

 reaction be recordedo 'Ibis field is defined by the following 

 methods by applying one's eye to the observation opening in 

 the canvas described above, the observer can bring the top 

 edge of this 2-inch square hole to coincide with the twine NO 

 (figo li;)> and the middle point of its right side with the 

 shadow of the electric light Lo With a fixed line and fixed 

 point so selected^, a square shown by the area PQRS in figure 

 lU is determinedo This area vms used as the field of observa= 

 tiono Since this area is immediately in front of the source 

 of light and away from the feeding site, and traversed by the 

 light beam^ entrance by tuna into this area as they travel 

 toward the light was taken as a criterion of reactiono Rec .rds 

 were made of the number of times during an experimental period 

 that the tuna swam into this fieldj and also of the pattern 

 of movemento An experimental period^, the time during which a 

 particular stimulus was applied ^^ was arbitrarily fixed at 5 

 minutes » 



Before starting a series of experiments;, preliminary obser- 

 vations were made of the behavior of the fisho liVhen more than 

 one was present they tended to form a school and cruLsed slowly 

 about the tank, occasionally passing between tlie observation 

 area and the baffle D (figo lU)s but more often circling in the 

 southern three-quarters of the tanko Only occasionally during 

 a period of about ^ minutes would tVey enter the observation 

 areas the number of entrances was usually Oj, occasionally Ij, 

 and more rarely 2 or 3o This was taken as control condition, 

 and care was taken that this condition was present before the 

 start of each experiment, although the number of entrances was 

 not always recorded nor included in the tables which follow. 

 At times the fish were unduly disturbed by extraneous factors 



U2 



