272 



Fishery Bulletin 94(2), 1996 



£><oo_J 



Figure 2 



Distribution of the four most abundant species of fish larvae 

 sampled with a rectangular midwater trawl (RMTl from 18 to 

 23 July 1991 on Emerald, Western, and Sable Island banks 

 (Scotian Shelf). 



Diet and feeding patterns in Atlantic 

 mackerel larvae 



The diet of Atlantic mackerel larvae changed mark- 

 edly during early ontogeny (Table 3). Small larvae 

 (<5 mm) fed almost exclusively on copepod nauplii. 

 With increasing length, their diet shifted from nau- 

 plii to copepods Icopepodites and adults) and fish lar- 

 vae. Copepod eggs did not contribute substantially 

 (<1% of the prey) to the diet of mackerel larvae <7 

 mm that preyed mainly on copepod nauplii. The in- 

 creased proportion of copepod eggs in the diet of lar- 

 vae >7 mm may have reflected the ingestion of egg- 

 bearing female copepods such as O. similis or 

 Pseudocalanus sp. Appendicularians were present in 

 small proportion (<5%) in the diet of all length classes 

 of larvae. The importance of fish larvae in the diet 

 increased with the length of Atlantic mackerel lar- 

 vae. The smallest Atlantic mackerel larva that had 

 preyed on another fish larva was 3.3 mm (fresh SL). 

 The mean cephalothorax length of the copepod prey 

 of Atlantic mackerel larvae increased threefold over 

 the range of lengths studied (<5 to >9 mm, Table 3). 

 Feeding incidence was high (>757<-) in all length 

 classes of mackerel. The feeding ratio (all prey types 

 considered) tended to decrease with length, a reflec- 

 tion of the increasing frequency of large prey such 

 as fish larvae. The incidence of predation on fish lar- 



vae increased with larval mackerel size, from 12% 

 in the smallest length class (<5 mm) to 69% in the 

 largest length class (>9 mm). 



The mean number of invertebrate prey in the gut 

 of Atlantic mackerel larvae was relatively constant 

 from the second half of the night to late afternoon 

 (1.9-3.5 prey per gut) and then jumped to 13 prey at 

 dusk or soon after (Fig. 4A). This sharp increase in 

 the number of invertebrate prey at dusk may repre- 

 sent an increase in copepod availability as the latter 

 pass through the layer containing mackerel larvae 

 during their nycthemeral migration towards the sur- 

 face. The percentage of dissected Atlantic mackerel 

 larvae with at least one fish larva in the gut tended 

 to increase from the morning (<30%) to the after- 

 noon and late evening (40-60%) (Fig. 4B). 



The gut of Atlantic mackerel larvae usually con- 

 tained no more than one fish larva (97% of cases). 

 Forty-four percent (314 out of 710) of mackerel lar- 

 vae containing a larval fish prey also contained at 

 least one invertebrate prey. Most fish larvae were 

 found in the midgut of Atlantic mackerel larvae 

 throughout the diurnal period (Fig. 5A). The fre- 

 quency of fish larvae in the foregut remained rela- 

 tively constant during daylight hours, increased at 

 dusk, and decreased in the second half of the night. 

 Occurrence in the reargut was maximum in the early 

 morning hours. In general, fish larvae in the gut of 

 mackerel larvae were either partly or well digested 

 (Fig. 5B). The percentage of Atlantic mackerel con- 

 taining slightly digested fish larvae was low and rela- 

 tively constant throughout the day The frequency of 

 partly digested fish larvae was high and tended to in- 

 crease from early morning to late afternoon, whereas 

 the frequency of well digested fish larvae declined. 



Prey selectivity in Atlantic mackerel larvae 



Atlantic mackerel larvae of all lengths fed selectively, 

 but selectivity for a given prey type changed during 

 ontogeny (Table 4). Selection for copepod nauplii was 

 high in small larvae (<5 mm) but declined rapidly as 

 predator size increased; larvae >7 mm selected against 

 this prey. Despite their numerical importance in the 

 diet, copepods were either selected against by larvae 

 <7 mm or consumed in rough proportion to their avail- 

 ability in the plankton by larger larvae. Selection for 

 fish larvae increased steadily with the length of Atlan- 

 tic mackerel larvae and preference for this prey type 

 was particularly strong in larvae >9 mm (Table 4). 



When analyzed by size categories, copepod prey 

 ( nauplii + copepodites + copepods ) <300 f.im in cepha- 

 lothorax length were selected (among the copepods) 

 by larvae <7 mm, whereas copepods in the range 300- 

 600 /.tm were selected against or consumed in near 



