FISHERY BULLETIN: VOL 86, NO 2 



the question posed, a 1-d study was thought 

 appropriate. Though environmental conditions 

 on this day might differ from "average", fish lar- 

 vae were assumed to be capable of a constant 

 array of behaviors. In other words, response (if 

 any) of the larvae to the vertical distribution of 

 their prey was assumed to be a deterministic 

 rather than a statistical phenomenon. If their 

 vertical distribution resembled that of their prey, 

 then food-seeking would remain a plausible ex- 

 planation for the near-bottom habit; if not, then 

 other stimuli must be considered important in 

 shaping these near-bottom concentrations of fish 

 larvae. 



The sampling was planned for daylight hours, 

 when most feeding by larvae was expected to 

 occur (Hunter 1981; Govoni et al. 1983). Late win- 

 ter was chosen because in this season peak larval 

 abundances of several species of interest to us 

 (northern anchovy, Engraulis mordax; white 

 croaker, Genyonemus lineatus; California hal- 

 ibut, Paralichthys californicus; and sometimes 

 queenfish, Seriphus politus, often overlap 

 (Lavenberg et al. 1986). A survey cruise in late 

 February found moderate-to-high abundance of 

 the first three species plus California sardine, 

 Sardinops sa^ox, (all >0.2 m"'^, Lavenberg un- 

 publ. data), and so this study was scheduled for 19 

 March 1985 off Seal Beach, CA (lat. 33°41'N, 

 long. 118°05'W; for a map, see Jahn and Laven- 

 berg 1986). 



As it happened, we chanced to encounter condi- 

 tions that were less typical than those found on 

 the February cruise. Only one fish species, white 

 croaker, was abundant enough to merit analysis, 

 and an uncommonly reported prey item, rotifera, 

 was important for small larvae. The diet of 

 various-sized larvae with respect to the abund- 

 ance of prey organisms at an array of heights 

 above the seabed was nevertheless useful in ques- 

 tioning whether food-seeking shaped the ob- 

 served larval distribution. 



METHODS 



Field 



At the hour of 0750 PST, an array of Interocean 

 model S4^ electromagnetic current meters was 

 set out over the 15 m isobath, with current meters 

 1, 4, and 8 m above the seabed. These meters were 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



set to record average current vectors and tem- 

 perature at 5-min intervals. The vessel (RV West- 

 wind) was then anchored some 200 m seaward of 

 the current meter array. A Nielson model NCH 

 fish pump, rated at 227 m'^ h^ at a 2 m head, was 

 used to sample fish larvae and zooplankton. The 

 end of the hose was tethered between a 200 kg flat 

 steel weight and several subsurface floats, with a 

 pulley arrangement such that divers could adjust 

 the distance between hose mouth and seabed. A 

 similar setup was previously found to give re- 

 peatable, fine-scale resolution at vertical sep- 

 arations of 25 cm (Jahn and Lavenberg 1986). 

 Sampling heights above the bottom were 50 cm, 

 1 m, and 6.7 m. The 15.2 cm diameter hose was 

 nearly horizontal at the tether point, so that 

 nominal sampling strata were z ± 7.6 cm. Ves- 

 sel surge, transmitted through the stiff hose, 

 caused occasional downward excursions of some 

 10 cm. 



Accompanying each pump sample was a cast of 

 water bottles for phytoplankton and microplank- 

 ton analysis. Rigid arrays of horizontally held 4 L 

 Niskin bottles (of. Owen 1981) were used to take 

 water samples simultaneously from 25, 50, and 

 100 cm above the bottom. The bottle array was 

 designed to be tripped by messenger, but poor 

 performance led to diver-implemented use after 

 the second cast. A midwater sample, 7.5 m below 

 the surface, was obtained via a single Niskin bot- 

 tle for each sample set. 



The sampling plan thus consisted of duplicate 

 pump samples from each of three strata, each 

 pump sample to be accompanied by a set of bottle 

 samples from four standard heights, three within 

 1 m of the seabed and one at midwater column. 

 One-liter samples from the bottles were fixed in 

 Lugol's solution for later identification of phyto- 

 plankton and microplankton. Pump samples of 

 15-min duration (approximately 35 m^) were 

 mainly directed into an overboard, 330 |xm mesh 

 plankton net for retention of large zooplankton 

 and ichthyoplankton. Unexpected problems in 

 reading an inline flowmeter required that vol- 

 umes be estimated as 2.4 m"^ min"^, based on pre- 

 vious experience with the pump under similar 

 conditions aboard the same vessel. To collect 

 smaller zooplankton, a 5 cm diameter hose led 

 from the intake side of the fish pump to a 100 jjtm 

 mesh plankton net. This small-meshed net was 

 suspended over a watertight box, which was 

 marked such that exactly 0.5 m'^ could be subsam- 

 pled for animals too small to be quantitatively 

 retained by the large net. This subsample, which 



252 



