FISHERY BULLETIN; VOL. 77, NO. 4 



mediately before feeding, dolphins were diving for 

 longer periods, perhaps going down deep to hunt 

 for food. This impression was not quantified in 

 direct association with feeding because we were 

 unable to identify particular individuals and thus 

 obtain length of dive records. However, we have 

 evidence from radio-tracked dusky dolphins indi- 

 cating that there is a length of dive-surface feed- 

 ing time association ( Wiirsig in press). Dive times 

 from six dusky dolphins radio-tracked in summer 

 showed a consistent increase in length during af- 

 ternoon, with average night and morning dives 

 about 14 s. Noon and afternoon dives rose to as 

 high as 32 s in average duration. One animal 

 radio-tracked in the austral winter (July-August 

 1974) showed no such length of dive increase in the 

 daytime and actually surfaced more frequently in 

 the afternoon than at other times of the day and 

 night. This is an indication that feeding in the 

 winter was different from feeding in the summer- 

 time. Since we believe that long dives in summer 

 are associated with surface feeding in deeper wa- 

 ter, it is likely that the extremely shallow and 

 brief dives which occurred at night (Wiirsig in 

 press) in summer were not associated with feed- 

 ing, and that perhaps the animals were resting 

 near the surface much of the night. This is the 

 reverse of what was found in the common dolphin 

 off California (Evans 1971, 1974), which dives for 

 long periods at night — and is believed to be feed- 

 ing at that time — and dives relatively shallowly 

 during the day. Once again, this difference may be 

 food related . While the common dolphin is thought 

 to feed upon the deep scattering layer which rises 

 out of deeper water enough for the dolphin to dive 

 to it at night, no defined deep scattering layer can 

 exist in the relatively shallow nearshore waters of 

 the present study (Hersey and Backus 1962). In- 

 stead, dolphins feed on anchovy during the day, 

 and move into deeper water as the day advances. 

 Whether or not anchovy move into deeper water 

 and are followed by the dolphin is not known. 

 Perhaps the daily movement into deeper water 

 was simply a consequence of being in shallow, 

 n'earshore water during night and early morning 

 (to be discussed later), and having to go into deeper 

 water in order to feed more efficiently. It is known 

 that individuals of southern anchovy schools dis- 

 perse during nighttime (Brandhorst and Castello 

 1971 ). This dispersal may make nighttime feeding 

 on anchovy more difficult or impossible, and there- 

 fore dolphins may rest at night while feeding dur- 

 ing the day in summer. 



As an app£u-ent consequence of feeding, dolphins 

 were also found more often in deep water in spring, 

 summer, and fall than in winter. Norris and Pres- 

 cott (1961) suggested a similar movement trend 

 ior Delphinus delphis in California waters. 



Group sizes were more often larger during the 

 surface feeding season. The reason for this was a 

 direct relationship between surface feeding activ- 

 ity and group size. Small groups usually engaged 

 in surface feeding for only brief periods. The 

 longer the feeding bout, the larger the number of 

 dolphins present. Dolphins appeared to begin feed- 

 ing in the morning and continued feeding through 

 most of the afternoon; thus, there was a general 

 increase in group size as the day advanced. The 

 many small groups in the morning (and presum- 

 ably night as well) covered a large (up to about 10 

 km in diameter) area, but nearest neighbors were 

 usually no more than 0.5 km apart. We assume 

 that they were probably within acoustic range of 

 each other. Why did surface feeding activity last 

 longer when dolphin numbers increased as groups 

 joined? Perhaps larger schools offish attract more 

 dolphins and keep them feeding for a longer time. 

 It is also possible that more dolphins are more 

 efficient at herding and maintaining the fish 

 school as a tightly clustered unit against the water 

 surface. As an alternative explanation, it might be 

 assumed that the small groups stopped feeding 

 after brief periods because individuals were 

 satiated. In a larger group, with perhaps more 

 individuals per fish school size and more competi- 

 tion, this would presumably take longer. Since 

 small schools which fed briefly were, however, 

 seen to feed more and more as the day advanced, it 

 seems unlikely that they had fed to satiation pre- 

 viously. Therefore, either larger fish schools sim- 

 ply attract more dolphins, or it is of direct advan- 

 tage to animals to feed in larger groups, and a 

 mechanism for telling nearby groups that herding 

 of foodfish is in progress may have evolved. Wari- 

 ous investigators have reported seasonal varia- 

 tions in group sizes, but none appear to link such 

 variations to a particular feeding mode as in the 

 present study. Gaskin (1972) stated that dusky 

 dolphins off New Zealand are found in smaller 

 schools in winter and larger ones in summer, basi- 

 cally the same as in our study. New Zealand dusky 

 dolphins feed on small squid and on surface fish, 

 but it is not clear whether their relative depen- 

 dence on these prey changes seasonally. 



How do other groups know about the feeding 

 bout 0.5-1.0 km distant? It is unlikely that at that 



884 



