FISHERY BULLETIN: VOL. 70, NO. 1 



not include any frames where fish were in single 

 file formation. 



Kawakawa does especially well in captivity 

 (Magnuson, 1965). Young adults were used, 

 of about 37 cm fork length; they had been in 

 holding tanks with running seawater for about 

 6 months, and appeared normal in all respects. 

 Thawed marine smelts (California osmerids), 

 were fed daily. The water temperature was 23 

 to 26° C, and was pumped into the tanks at 520 

 liters/min from a storage sump fed by a well 

 6 to 9 m deep. The salinity was 33 to 35;^r. Dis- 

 solved oxygen levels averaged 6 mg/liter. A 

 1,000 watt and a 300 watt underwater incan- 

 descent light, at the bottom of the tank, were 

 used, in addition to banks of overhead fluor- 

 escents, to simulate daylight. 



RESULTS 



In the absence of partitions the diagonal to 

 abeam position ratios for fish orienting in par- 

 allel were 5.17/1 and 6.04/1 respectively in the 

 rectangular and octagonal areas. It appeared 

 that the abeam position was assumed for only 

 brief periods when changing speeds in the course 

 of altering direction, turning, or shifting posi- 

 tion in the school. In the presence of partitions 

 the ratios were 1.28/1 and 2.36/1 respectively 

 in the rectangular and octagonal areas. Thus 

 the abeam position assumed increased impor- 

 tance during parallel orientation with partitions 

 (Table 1). 



The side-to-side spacing is shown in Table 2 

 as the mean distance apart of either the two fish 

 in a group of two, or of the two closest fish in a 

 group of three or more. Each mean represents 

 measurements made between a single pair of 

 fish during a single continuous film sequence. 

 Without partitions, this distance ranged from 

 14.0 to 29.8 cm in the rectangular area, and 13.4 

 to 23.0 cm in the octagonal area. Fish separated 

 by a partition stayed much farther apart; the 

 side-to-side spacing ranged from 64 to 134 cm 

 in the rectangular area (Figure 3) , and from 30 

 to 82 cm in the octagonal area. Fish-to-partition 

 spacing varied from a minimum of about 7 to 

 8 cm to a maximum of about 75 cm, and appeared 

 to depend on the location of the other fish (Fig- 



Table 1. — Positional orientation of schooling Euthynnus. 



' These are equal to the number of frames measured for each category 

 in Table 2. 



Table 2. — Side-to-side spacing of Euthynmis in schooling 



orientation. 



' This partition contained the series of holes. 

 - There were only 4 fish in this group. 



ure 4). Circular holes in the Plexiglas did not 

 change the spacing (Table 2). Water movement 

 through the holes was very slow. Fifteen min- 

 utes after fluorescein dye was placed on one side 



200 



