250 



Fishery Bulletin 93(2), 1995 



and Dunn, 1985). In this region, fish eggs in the neus- 

 ton could be considered strays because their accu- 

 mulation at the surface may be attributed to their 

 positive buoyancy rather than to the deposition of 

 eggs in this zone. A similar conclusion has been made 

 regarding the occurrence of fish eggs in the neuston 

 of shelf and oceanic waters off Washington, Oregon, 

 and northern California (Doyle, 1992). 



In contrast, a unique group of larval fish appeared 

 to be associated with the neuston in the western Gulf 

 of Alaska, and most of the dominant taxa were scarce 

 or absent in the subsurface zone. The dominance of 

 hexagrammids, cottids, an osmerid, Arcop/opoma/ira- 

 bria, bathymasterids, Cryptacanthodes aleutensis, 

 and Ammodytes hexapterus in this group has also 

 been documented for the larval fish component of the 

 neuston in the California Current region along the 

 U.S. west coast (Brodeur et al., 1987; Shenker, 1988; 

 Doyle, 1992). The occurrence of T. chalcogramma 

 larvae in high numbers in the neuston of the west- 

 ern Gulf of Alaska, however, is unique to this region 

 and reflects the overall dominance of this species in 

 the plankton of the study area (Kendall and Dunn, 

 1985; Schumacher and Kendall, 1991; Rugen 3 ). 



Among the dominant taxa of fish larvae in the 

 neuston, most were obligate tenants of the surface, 

 despite the predominance of demersal spawning 

 among these taxa. The most important taxa in this 

 group included the hexagrammids, cottids, Anoplopoma 

 fimbria, and Cryptacanthodes aleutensis, and, accord- 

 ing to the general classification scheme for neustonic 

 organisms (Zaitsev, 1970; Hempel and Weikert, 1972; 

 Peres, 1982), they may be considered obligate mem- 

 bers of the neuston. The same taxa of larvae have 

 been identified as obligate neustonic organisms in 

 the plankton off the U.S. west coast (Doyle, 1992). 

 Because of their scarcity in bongonet samples, the 

 dramatic daytime reduction in density of these lar- 

 vae in the neuston samples may have been attrib- 

 uted primarily to light-aided avoidance of the sam- 

 pling gear. The generally large sizes documented (pre- 

 dominantly > 10 mm SL) for these neustonic larvae also 

 contributed to their ability to avoid the neuston net. 



Theragra chalcogramma, Ammodytes hexapterus, 

 and Bathymaster spp. larvae were unusual among 

 the dominant neustonic taxa in that they were ex- 

 tremely abundant in bongo net samples also; there- 

 fore, their association with the neuston was consid- 

 ered facultative. Their nighttime presence in the 

 neuston suggested a pattern of diel vertical migra- 

 tion with movement upward at dusk and a return to 

 deeper layers during the day. This pattern has been 

 observed for many species of fish larvae and zoo- 

 plankton in many different regions (Zaitsev, 1970; 

 Hempel and Weikert, 1972; Neilson and Perry, 1990). 



Mallotus villosus, Myoxocephalus spp., and Zaprora 

 silenus larvae, which were well represented in bongo 

 net samples, but abundant in the neuston at night, 

 may also exhibit this pattern of vertical migration. 

 However, with the limited data presented here, it 

 was difficult to verify this migration pattern. Day- 

 time sampler avoidance, particularly by the larger 

 larvae and early juveniles, is likely to have had a 

 confounding influence on the observation of such a 

 pattern. In addition, it is necessary to consider in 

 more detail the diel variation in the vertical distri- 

 bution pattern of the larvae over the entire range of 

 the water column in which they occurred. 



Kendall et al. (1987, 1994) observed that within 

 the upper 50 m of the water column, T. chalcogramma 

 larvae (size range approximately 7-10 mm SL) un- 

 dergo limited vertical migration on a diel cycle. These 

 larvae were found to be deepest during the day, shal- 

 lowest in the evening, sink slightly at night, and sink 

 more in the morning. Under controlled laboratory 

 conditions, Olla and Davis ( 1990) also observed simi- 

 lar diel periodicity in vertical distribution of T. chal- 

 cogramma larvae; larvae moved downward with day- 

 time light intensity, upward during evening twilight 

 conditions, remained close to the surface at night, 

 and moved downward again in the morning. The T. 

 chalcogramma larvae caught in the neuston, mainly 

 at night, during the present study were predomi- 

 nantly 5-14 mm SL unlike their counterparts in the 

 bongo net samples that were mostly <6 mm SL. This 

 diel pattern of neustonic occurrence for the larger- 

 sized larvae was likely due to the pattern of diel ver- 

 tical migration observed by the above authors. 



Observations on the vertical distribution of Ammo- 

 dytes hexapterus larvae have also been made in the 

 western Gulf of Alaska (Rogers et al., 1979; Brodeur 

 and Rugen, 1994). Unlike T. chalcogramma larvae, 

 A. hexapterus larvae were found to be deepest in the 

 water column at night and shallowest at dawn and 

 during the day. This apparent migration pattern of 

 nocturnal descent has also been observed for A. 

 personatus larvae off Japan and has been interpreted 

 as advantageous in terms of diurnal feeding and 

 predator avoidance (Yamashita et al., 1985). If this 

 is the normal diel pattern of vertical migration for 

 Ammodytes larvae, the occurrence of high densities 

 of A. hexapterus larvae in the neuston at night, docu- 

 mented during the present study, seems unusual. On 

 examination of length-frequency distributions for 

 these larvae, however, it appears that the pattern of 

 nocturnal descent was prevalent among larvae <20 

 mm SL (Yamashita et al., 1985; Brodeur and Rugen, 

 1994), whereas the nocturnal concentration of lar- 

 vae at the surface was restricted to larger larvae and 

 early juveniles (Doyle, 1992; this study). Perhaps 



