294 



Fishery Bulletin 91(2). 1993 



dock larvae (<9mml below the thermocline, we found 

 both haddock and cod larvae of this size tended to 

 reside mostly within or above the base of the thermo- 

 cline in May on Georges Bank. Miller et al.'s (1963) 

 ability to discriminate depth changes in larvae was 

 limited because of the small, non-opening/closing sam- 

 plers ( 14 cm mouth diameter) that were towed at fixed 

 depth intervals (minimum 9-10 m apart). 



Ellertsen et al. (1984) reported on the vertical 

 distribution of cod larvae (4-5 mm) at a strongly- 

 stratified station off the Lofoten Islands, Norway, in 

 May 1982. First-feeding cod larvae were concentrated 

 at 10-20 m depth near the thermocline throughout a 

 24 h period, but dispersed with the slightest wind mix- 

 ing. Only under extremely calm conditions were the 

 larvae able to control their vertical distribution and 

 show diurnal migration. 



Pelagic juveniles, however, are capable of making 

 extensive migrations through the water column. In our 

 Georges Bank studies, at a well-mixed shoal site in 

 June 1984, pelagic cod juveniles appeared to migrate 

 into the upper half of the water column at night and 

 to remain in the lower water column during the day. 

 Juvenile haddock were much less abundant at this 

 site, and their center of abundance was located at mid- 

 depth with no evidence of day-night migrations like 

 those of cod. The few pelagic cod collected at two strati- 

 fied sites on eastern Georges Bank in July 1985 were 

 generally found in the lower third of the water col- 

 umn, and few were caught in the thermocline at 15- 

 30 m depth. We have no data on pelagic juvenile had- 

 dock at a stratified site for comparison. 



Perry & Neilson ( 1988) studied the vertical distribu- 

 tion of cod and haddock pelagic juveniles in June 1985 

 on eastern Georges Bank and found a diel migration 

 pattern for both species at a well-mixed shoal site 

 (66 m). At a nearby stratified site in deeper water 

 (80 m), Perry & Neilson (1988) found that juvenile had- 

 dock generally remained above 40 m depth with their 

 weighted-mean depth in the thermocline at 20-30 m. 

 In contrast, cod juveniles generally remained deeper 

 than 40 m, below the thermocline, yet still exhibited 

 diel migrations. 



Colton ( 1965) investigated vertical distribution of ju- 

 venile haddock using an Isaacs-Kidd midwater trawl 

 in late summer of 1957-58. Over 75% of the pelagic 

 haddock of 27-124 mm occurred between 10 and 40 m 

 depth, with greatest abundance at 20 m at the depth 

 of the thermocline. There was some diel variation in 

 distribution: the weighted-mean abundance was lo- 

 cated at 40 m during the day and 30 m at night. 



Results from our studies indicate that juvenile had- 

 dock and cod change from a pelagic to a demersal 

 existence at a body length of 4-10 cm, and once they 

 reach a demersal stage they stay close to bottom by 



day and move off bottom into the water column at 

 night. During the transition period the vertical ampli- 

 tude of these night excursions decreases with size of 

 fish. The data suggest that cod <6cm make more ex- 

 tensive off-bottom migrations at night than haddock, 

 with some individuals ranging up to the surface. Had- 

 dock appear to assume a more complete demersal life 

 at a smaller size (4-6 cm) than cod, which is consis- 

 tent with older juvenile and adult behavior. Bailey 

 (1975) found that 6.5-13 cm haddock migrated through 

 the water column at night, whereas larger fish were 

 more confined to the seabed in the northern North Sea 

 during August. Beamish (1966) and Woodhead (1966) 

 reported that adult haddock tend to remain associated 

 with the seabed at night, whereas adult cod make ex- 

 tensive vertical migrations. In contrast, Bailey (1975) 

 concluded that cod juveniles remained on or close to 

 the seabed day and night, because fish caught in 

 midwater were never >7cm, while those on the seabed 

 ranged up to 13 cm. 



Juvenile cod and haddock at -70-90 mm begin feed- 

 ing on benthic prey such as polychaetes and crusta- 

 ceans along with planktonic prey, which is consistent 

 with changes in their mouth (Auditore et al. 1993). 

 The time for juvenile cod and haddock to grow from 

 40 mm to 70 mm is -30 d, and from 40 to 90 mm is 

 -48 d (Bolz & Lough 1988). Therefore, in terms of de- 

 fining the functional transition period for an individual 

 fish, we consider the transition from a wholly pelagic 

 to demersal existence to take -1-1.5 mo on average. 



Other factors influencing vertical 

 distributions 



Neilson & Perry (1990) did a recent review of the lit- 

 erature on diel vertical migrations of marine fishes 

 and concluded that the diversity of vertical migration 

 patterns indicated a facultative process significantly 

 influenced by multiple environmental factors. Prey dis- 

 tributions probably play an important role in the ver- 

 tical distributions of larvae. Based on Georges Bank 

 data (Buckley & Lough 1987) and simulation models 

 (Laurence 1985), haddock larvae are considerably more 

 food-limited than cod. Stable stratified conditions in 

 the spring, which result in the availability of high con- 

 centrations of zooplankton prey, favor growth and sur- 

 vival of haddock larvae. Lough (1984) and Buckley & 

 Lough (1987) found that haddock and cod larvae 

 <~13mm (mode at 7mm) were generally associated 

 with the planktonic prey biomass that had a maxima 

 in the thermocline during May 1981 and 1983 along 

 the southern margin of Georges Bank. The stronger 

 the water column stratification, the more closely lar- 

 vae and prey coincided. At the well-mixed sites where 

 larvae were broadly distributed, prey biomass was 



