Fishery Bulletin 88(4). 1990 



Median widths of diatoms were the smallest (0.01-0.10 

 mm), followed by copepod eggs, nauplii, and 0. similis 

 (0.11-0.20 mm), Pseudocalanus copepodites, A. longi- 

 remis, Acartia spp., unidentified copepods, and other 

 prey (0.21-0.30 mm), and Pseudocalanus sp. adults and 

 amphipods (0.31-0.40 mm). 



Results 



A single prey-size class was most often ingested by all 

 larvae (Fig. 2A). While both large and small larvae con- 

 sumed small prey, the ingestion of larger prey by large 

 larvae resulted in the two curves being significantly 

 different (P<0.01, Kolmogorov-Smirnov test; Conover 

 1980). The shared mode of the two curves was attrib- 

 uted to large larvae ingesting a number of small prey 

 items, particularly copepod eggs (0.11-0.20 mm) (Table 

 1). When we eliminated copepod eggs from the anal- 

 yses, assuming that they were taken incidentally in the 

 course of ingesting adult female Pseudocalanus sp., the 

 prey-size distribution of small larvae did not change 

 noticeably (Fig. 2B). But the removal of copepod eggs 

 from the diet of large larvae shifted the prey-size mode 

 towards larger prey. 



Copepod nauplii were the primary prey of small lar- 

 vae (Table 1). Pseudocalanus sp. also comprised a major 

 portion of the diet, with adults and copepodites nearly 

 equal in importance. Unidentified copepods made a 

 small contribution, while Acariia spp., Oithona similis, 

 and copepod eggs were rather insignificant in the diet 

 of small larvae. 



Copepods were the primary prey of large larvae 

 (Table 1), with Pseudocalanus sp. adults predominant 

 over copepodites and all other species. A large number 

 of copepod eggs were ingested, while amphipods, un- 

 identified copepods, Acartia longiremis, and copepod 

 nauplii made progressively smaller contributions to the 

 diet. The relatively large size of amphipods likely con- 

 tributed to their being ingested only by large larvae. 



The mean number of prey items ingested by small 

 larvae was 26.9, compared with 34.1 by large larvae. 

 The incidence of feeding was 100%. 



Discussion 



The pattern of prey-size selection observed in the Ber- 

 ing Sea contrasts with the pattern seen for larvae col- 

 lected off the Oregon and Washington coasts in 1980 

 (Grover and 011a 1986). In each study functional breaks 

 in diet (e.g., the shift away from dependence on cope- 

 pod nauplii) were used to define larval size groups. The 

 Oregon- Washington collections were categorized in- 

 to three size groups: 8.2-12.5, 12.6-20.5, and 20.6-28.5 



o 



Q_ 



O 

 O 



o 



80 

 70 ~ 

 60- 

 50- 

 40- 

 30- 

 20 

 10 

 



16 

 12 



23 mm SL 



1 5 mm SL 



001- oil- OPI- 

 OID 20 OJO 



OBI- 



90 



PREY WIDTH (mm) 



Figure 2 



Size of prey selected by two size classes of larval sablefish (12-15 

 and 16-2.3 mm SL) in the Bering Sea in 1979: (A) including copepod 

 eggs, (B) omitting copepod eggs. 



mm SL. Each size group favored a different peak prey 

 size, with the smallest larvae ingesting the smallest 

 prey, and the largest larvae ingesting the largest prey 

 (Grover and 011a 1986). However, peak prey sizes for 

 the two size groups from the Bering Sea were the same 

 as long as copepod eggs were included in the analyses. 

 Their removal shifted the prey-size utilization curve of 

 large larvae in the Bering Sea towards larger prey. 

 Copepod eggs had a greater impact on the diet in the 

 Bering Sea than off Oregon due to a difference in the 

 biology of the dominant copepod species ingested in the 

 two regions. The dominant copepod in the diet of large 

 larvae collected off Oregon in 1980 was Paracalanus 

 parvus (Grover and OUa 1987), a species that broad- 

 casts its eggs into the sea (Checkley 1980). Pseudo- 

 calanus sp., the dominant copepod in the diet of larvae 

 collected from the Bering Sea, carries its eggs (Cor- 

 kett and McLaren 1978) which could thus be ingested 

 incidental to the adults. Nevertheless, as female cope- 

 pods with eggs were not observed in sablefish guts, and 

 as other copepods that broadcast their eggs (such as 

 all Acartia spp.) contributed to the diet, it remains a 

 possibility that some eggs were ingested independently 



