208 



Fishery Bulletin 104(2) 



numerically important in the area (Pepin an(i Maillet''), 

 but O'Driscoll et al. (2001) did not find this species in 

 any of their stomach samples despite its importance in 

 the regional plankton community. Fish eggs represented 

 an average of 0.13% of the total zooplankton community 

 among all sites sampled (Table 1). 



Only 12 of 218 capelin stomachs, for which com- 

 plete stomach analysis was performed, were found 

 to be empty. The weight of prey in capelin stomachs 

 increased with increasing length but the number of 

 prey decreased with increasing length, indicating that 

 the average size of prey increased with capelin length 

 (Fig. 1). Stomach fullness increased at sunset, peaked 

 shortly after midnight, and then gradually decreased 

 during the remainder of the night and day (Fig. 1). 

 Copepods were the only prey in 73 of the remaining 

 206 capelin. Euphausiids and amphipods were the next 

 most abundant prey item, found in approximately 25% 

 of capelin stomachs and representing on average 6.5% 

 and 8.8% of prey by number (Table 2). Pteropods, larva- 

 ceans, cladocerans and fish eggs were present in approx- 

 imately 10% of capelin stomachs and represented -1-3% 

 of prey by numbers (Table 2). Other prey taxa were 

 found in a few capelin stomachs. Principal components 

 analysis showed that the proportion of euphausiids 



^ Pepin, P.. and G. L. Maillet. 2002. Biological and chemi- 

 cal oceanographic conditions on the Newfoundland Shelf 

 during 2001 with comparisons with earlier observations. 

 Canadian Science Advisory Secretariat, Research Document 

 2001/073, 60 p. [Available from http;//www.dfo-mpo.gc.ca/ 

 csas/ or Canadian Science Advisory Secretariat, Depart- 

 ment of Fisheries and Oceans, 200 Kent Street, stn. 12032, 

 Ottawa, Ontario, Canada KIA 0E6.J 



and amphipods in stomachs increased with capelin 

 length, whereas copepods were more frequent in small- 

 er fish, as indicated by the first principal component 

 (Table 3). Other prey categories, particularly fish eggs 

 and larvaceans, were generally more likely to occur in 

 capelin stomachs when they were relatively empty, as 

 indicated by the second principle component (Table 3). 

 Finally, the third principal component indicated that 

 fish eggs appeared to be less important in areas where 

 mollusks and Cladocera were more frequent in the diet 

 of capelin — a finding that possibly reflects the influence 

 of water masses in which coastal rather than shelf zoo- 

 plankton were dominant (Table 3). These observations 

 of prey composition were generally consistent with those 

 of O'Driscoll et al. (2001). 



When we consider all stomachs analyzed for fish eggs, 

 the overall frequency of occurrence of fish eggs was 

 4.3%- (46 of 1052 stomachs). This is less than the value 

 (7.9% of stomachs with prey) obtained from those stom- 

 achs with complete analysis of prey composition. The 

 frequency distribution of fish eggs was well described 

 by a Poisson distribution (Fig. 2). The majority offish 

 eggs (89%) were of the CHW complex and the remainder 

 (11%) were of the CYT complex. The generalized linear 

 model revealed that there was a significant effect of 

 time of day (Table 4), and the greatest number of fish 

 eggs in capelin stomachs were found between 20:00 

 and 24:00, following sunset, and the lowest numbers 

 were found between 00:00-04:00 (Fig. 3). There was 

 approximately a 20-fold difference in the least squares 



