Lough and Potter Vertical distribution of Melanogrammus aeglefinus and Gadus morhua 



295 



lower and more uniformly distributed through the wa- 

 ter column. 



Frank & McRuer ( 1989) also found the condition of 

 haddock larvae off Southwestern Nova Scotia in May 

 1986 consistent with the well-mixed/stratified regime 

 findings of Buckley & Lough (1987). Larvae were in 

 good condition at the deeper stratified sites and in 

 poor condition at the shallow, well-mixed sites. They 

 also found evidence that the buoyancy of starved lar- 

 vae (4— 7 mm) increased, which frequently resulted in 

 a bimodal depth profile, with larvae in poorer condi- 

 tion nearer the surface. Ellertsen et al.'s (1984) study 

 in the Vesteralsfjord of Norway showed that under 

 calm conditions larval cod appeared to concentrate in 

 the surface layers during the night coinciding with the 

 highest density of copepod nauplii, their preferred prey 

 in the study. They also noted that turbulent mixing 

 dispersed the larvae and prey evenly throughout the 

 water column. Therefore, the vertical distribution of 

 larvae and their prey needs to be evaluated in relation 

 to larval buoyancy and water-column turbulence. 



Light plays an important role in the diel migrations 

 of cod and haddock larvae, but it is not known whether 

 larvae follow a specific isolume. There may be a criti- 

 cal light intensity at which visual feeding ceases and 

 vertical migration begins. Minimum light-intensity 

 threshold for larval cod feeding was in the range 0.1- 

 0.4 lux, and maximum feeding incidence at 1.4 lux 

 (Ellertsen et al. 1980). Light regime also can affect the 

 swimming activity and buoyancy of yolksac cod larvae 

 by changing their metabolism and utilization of yolk 

 (Solberg&Tilseth 1987). 



In Perry & Neilson's (1988) study of pelagic juve- 

 niles, the migration pattern of cod appeared to be di- 

 rectly related to the light-dark cycle, whereas that of 

 haddock was more complicated and may have been 

 confounded by periodic changes in the rotary tidal cur- 

 rent speed and/or the migratory movements of their 

 preferred mysid prey Neomysis americana. Their ex- 

 planations for the different depth distribution may be 

 related to (1) cod's acclimation to colder waters and 

 haddock's preference for warmer waters, and (2) a pos- 

 sible mechanism for reducing interspecific competition, 

 since their vertical distribution patterns were similar 

 when prey biomass was high and different when prey 

 biomass was low. In our opinion, the differences in 

 vertical distribution between cod and haddock at the 

 stratified site noted by Perry & Neilson (1988) may be 

 attributed to species differences in feeding behavior. 



On Georges Bank, planktonic prey selected by both 

 species during settlement at 40-60 mm include copep- 

 ods, amphipods, euphausiids, and cumaceans (Bow- 

 man 1981ab, Koeller et al. 1986, Mahon & Neilson 

 1987, Lough et al. 1989). Robb (1981) showed that 

 pelagic 0-group gadids in the northern North Sea have 



a different feeding behavior reflective of the adult 

 lifestyle, i.e., cod fed on larger active pelagic prey, while 

 haddock concentrated on smaller slower-moving or sed- 

 entary organisms. Both Robb's (1981) and Perry & 

 Neilson's (1988) feeding results support the view that 

 haddock feed continuously when prey is available. 



A complicating factor in understanding vertical dis- 

 tribution patterns in haddock is their commensal as- 

 sociation with the large coelenterate Cyanea sp. This 

 association has been reported in the literature for fish 

 up to a length of ~10cm (Colton & Temple 1961, 

 Mansueti 1963, Rees 1966, Bailey 1975, Koeller et al. 

 1986). Mansueti (1963) stated that jellyfish serve as a 

 source of food and shelter for young fish between the 

 pelagic and demersal stages. The change from a sur- 

 face to deeper-water habitat by the fish is believed by 

 Mansueti (1963) to be a factor causing the breakup of 

 the commensal relationship. Koeller et al. (1986) 

 showed Cyanea sp. to have been most abundant at the 

 surface, decreasing to near zero below 80 m off Nova 

 Scotia in June 1983. Pelagic juvenile haddock, more so 

 than cod, may require contact with some objects such 

 as jellyfish. Whether jellyfish tentacles also serve as a 

 source of food is not known. 



The substrate on which the juveniles settle, (i.e., an 

 unsuitable habitat such as sand) may prolong the tran- 

 sition to demersal life. Lough et al. (1989) hypoth- 

 esized that a large gravel habitat on northeastern 

 Georges Bank favors survival of recently-settled juve- 

 nile fish because of predator avoidance through cam- 

 ouflage and/or increased prey abundance. Pelagic ju- 

 venile cod were widespread over eastern Georges Bank 

 in June; and in mid-July when they became demersally 

 oriented, they were present on bottom types ranging 

 from sand to gravel. By contrast, in late July to early 

 August, recently-settled juveniles were abundant only 

 on the large gravel deposit on northeastern Georges 

 Bank and were sparse or absent from the large sand 

 and gravelly sand that cover most of eastern Georges 

 Bank (Lough et al. 1989). 



To summarize the vertical distribution patterns ob- 

 served, we developed a generalized day-night mean 

 depth distribution of cod by size-class, from recently- 

 hatched larvae through recently-settled juveniles 

 (Fig. 20). Cod vertical migrations appear to respond 

 primarily to a day-night cycle, whereas haddock mi- 

 grations respond to other factors such as prey distri- 

 bution. Depth centers of the smallest larvae 

 (2-8 mm) appear at two levels-15 and 35 m-depend- 

 ing on whether the water column is stratified or well- 

 mixed (Fig. 20). In order for the water column to be 

 considered sufficiently stratified for larvae to initially 

 reside at 15 m, a change of >0.3 sigma-< units would 

 have to occur from surface to the base of a pycnocline. 

 Waters around the flank of Georges Bank are typically 



