FISHERY BULLETIN: VOL. 75, NO. 1 



1971 



1972 



| > 70 00 



15000-6999 [33000-4999 D<30 00 



FIGURE 4. — Station to station similarity-coefficient matrices for 1971 and 1972 data on larval fishes based on Sander's (1960) 

 dominance affinity index. All taxa except Sebastes spp. were included in the analysis. 



excluded from the analysis to minimize masking 

 effects that might have arisen because of the 

 multispecies nature of the group. Since osmerids 

 were known to be essentially coastal forms, they 

 were not excluded. 



Peaks in larval abundances were associated 

 with the location of these two assemblages with an 

 apparent transitional zone of low larval abun- 

 dance between them (Figure 5). In both 1971 and 

 1972 abundance was relatively high inshore, 

 dropped to a low at 28 km, and then increased 

 seaward. 



Larval taxa were determined to be associated 

 with the coastal or offshore zone on the basis of 

 whether 80% or more of all larvae were taken at 

 stations 2 to 28 (coastal = C) or stations 37 to 111 

 (offshore = O). Using these criteria, 84 of the 90 

 taxa (93%) could be designated as coastal or 

 offshore (Table 2). Fifty-three taxa in 16 families 

 and 1 order were coastal. Of these, 49 were 

 identified to species, 3 to family, and 1 to order. 

 Ninety-nine percent of all larvae in these 53 taxa 

 were taken in the coastal zone 2 to 28 km offshore. 

 Thirty-one taxa in 15 families were offshore. Of 

 these, 26 were identified to species, 4 to genus, and 

 1 to family. Ninety-six percent of all larvae in 

 these 31 taxa were taken 37 to 111 km offshore. 



Only six taxa could not be designated as coastal 

 or offshore. This was probably due in part to rarity, 



e.g., Hippoglossoides elassodon (total standard- 

 ized number = 5.29; 51% were C and 49% were 

 0),Bathyagonus spp. (3.30; 47% C and 53% O), and 

 to multispecies groups, e.g., Cyclopteridae spp. 1 

 (30% C and 70% O) and Bathyagonus spp. In- 

 terestingly, 96% of all Sebastes spp. larvae were 

 taken in the offshore area. Leptocottus armatus 

 was primarily coastal since 77% of all larvae were 

 taken there. Only one sample outside the coastal 

 area (Station 37, in February 1971) contained L. 

 armatus larvae, but they were present in moder- 

 ate numbers. Hemilepidotus hemilepidotus (67% C 

 and 33% O) and//, spinosus {IWc C and 30% O) 

 distributions are more difficult to explain. 

 Hemilepidotus spinosus larvae in the coastal area 

 were smaller (4-9 mm, x 5.3) than those farther 

 offshore (6-12 mm, x 8.9) as were//, hemilepidotus 

 (4-6 mm,x 5.2 in the coastal area and 8-11 mm, x 

 9.3 offshore). Hemilepidotus spinosus larvae are 

 sometimes abundant (>600 larvae/15 min tow) in 

 the neuston (upper 15 cm of the water column), 

 particularly at night (Richardson unpubl. data). 

 These data suggest that larvae which are as- 

 sociated with surface waters may undergo some 

 kind of offshore transport which does not affect 

 nonneustonic species. 



Modes of reproduction differ considerably 

 between those species designated as coastal and 

 those designated as offshore. Of the 53 coastal taxa 



132 



