FISHERY BULLETIN: VOL. 70, NO. 3 



ET Sec 0.000 



0.117 



0.359 



0.601 



0.842 



123 



1.427 



.435 



1.443 



1.451 



1.466 



.498 



.576 



.794 



Figure 5. — Tracings of selected motion picture frames from a feeding sequence of an anchovy larva (8.9 mm, 

 21 days old) and a prey taken at 128 frames per second. Frame lines are fixed such that distance moved by larva 

 is indicated by comparison of tracings; and unshaded image indicates position of larva in the preceding tracing. 

 The following events are illustrated: elapsed time (ET) sec, larva sights prey; ET 0-0.117 sec, larva orients 

 head toward prey and swims toward it; ET 0.117-1.435 sec, larva forms s-shaped strike posture; ET 1.435-1.443 

 sec, strike begins and prey captured; and ET 1.451-1.794 sec, forward movement continues as tail returns to axis 

 of progression. 



tude of the posterior bend in the body reached 

 a maximum just before the strike was begun. 



The larva began the strike by drawing the 

 tip of the tail rapidly anteriorly. It then thrust 

 the tail backward, extended the head, opened the 

 mouth, and the particle was taken or missed 

 within 7.8 to 15.6 msec after the beginning of 

 the backward thrust of the tail (one to two 

 frames at 128 fps) . The entire feeding sequence 

 from sighting the prey to the strike lasted only 

 1 to 2 sec. The sequence of movements were so 

 rapid that my first impression was that they 

 were extremely stereotyped. Later, frame-by- 

 frame analysis of film taken at 128 fps indicated 

 that normal variance was associated with every 

 characteristic that could be measured in the films. 



Larvae of 15 to 20 mm formed the strike pos- 

 ture in less time than did one 5 to 15 mm long, 

 and consequently, feeding sequences were short- 

 er in larger larvae (Figure 6). On the other 

 hand, the time used to approach the prey before 

 forming the strike posture was about 0.6 sec in 

 larvae of all sizes. Other characteristics in- 

 creased directly in proportion to length, for ex- 

 ample, the maximum amplitude of the body in 

 the strike posture, the distance to the prey at the 

 time of the strike, and the rate of movement 

 toward the prey while in the strike posture. The 

 speed of the strike also increased with length 

 but the relationship appears to be nonlinear. 

 The speed of the strike relative to length was 

 less in the largest size class of larvae. 



828 



