406 



Fishery Bulletin 98(2) 



increasing the foraging area (Robinson and Pitcher, 

 1989; Robinson. 1995). 



Age-1 Pacific herring repeated this contraction- 

 expansion pattern of school structure and distri- 

 bution during the year. After their second winter, 

 herring appeared to join the adult schools, leaving 

 as new recruits entered the bays. 



Pacific herring and walleye pollock geographic dis- 

 tributions were roughly similar within Prince Wil- 

 liam Sound. Walleye pollock distribution was bimodal 

 with aggregations in the east-northeast and the west- 

 southwest areas. However, there appeared to be little 

 spatial overlap between these species because they 

 occupied different portions of the water column. Juve- 

 nile walleye pollock were highly aggregated, forming 

 a few, very dense schools in July within bays. Juve- 



nile walleye pollock distribution may be influenced 

 by the cannibalistic behavior of adult walleye pollock 

 (Livingston, 1993). Adult walleye pollock were pri- 

 marily aggregated in the southern portion of Prince 

 William Sound, although a smaller school was evi- 

 dent in the east-northeast. 



Bays within Prince William Sound appear to be 

 nursery areas for both Pacific herring and walleye 

 pollock juveniles. Data presented here and growth 

 studies suggest that these nursery areas are iso- 

 lated (Stokesbury et al., 1999). This finding differs 

 from that for other herring species distributions 

 where juveniles from different populations appear 

 to share a common nursery and feeding areas, sep- 

 arating only when adults begin to spawn (Blaxter 

 and Hunter, 1982; lies and Sinclair, 1982; Sinclair et 



