OWEN ET AL.: SPAWNING AND SURVIVAL OF LARVAL ANCHOVY 



Environmental Characteristics 



Surface Drift 



Drifter positions were determined by a com- 

 bination of satellite navigation, Loran-C, and 

 Decca radar over the com'se of the study, and 

 were precise within 0.1 km. The drifters were 

 designed as described by Davis et al. (1982) to 

 minimize the effects of wave motion and of wind 

 drag on exposed surfaces, both of which cause 

 deviations of drifter motion from that of water 

 parcels. The 2 km station spacing around each 

 drifter was large compared with precision of 

 drifter position and with deviation of the 

 drifter's motion from surrounding water 

 motion. 



Vertical Temperature and Salinity Profiles 



Temperature, conductivity, and pressure were 

 recorded with a Neil Brown CTD, following 

 methods used by NOAA/Southwest Fisheries 

 Center.' Briefly, 0.25 s scans of temperature, 

 conductivity, and pressure were recorded as the 

 CTD profiled at 20-40 m/min. Conductivity and 

 temperature records corrected for temperature 

 lags are smoothed by a 5-point weighted running 

 mean with binomial coefficients of 1, 4, 6, 4, 1. 

 Sahnity was computed using the Practical Salin- 

 ity Scale of 1978. Salinities and temperature 

 were compared and justified with salinity and 

 temperature values from a hydrogi-aphic bottle 

 tripped in the near-surface layer during each 

 cast. Final data were enumerated at 2 m depth 

 intervals. 



Phytoplankton Pigments 



Chlorophyll-a and phaeopigment-a concentra- 

 tions were determined from fluorescence read- 

 ings on a Turner 111" Fluorometer before and 

 after acidification of 24 h extractions in 90% 

 aqueous acetone of material retained on What- 

 man GF/C glass filters after filtration of 140 mL 

 water samples. 



Particles 



Particle concentration and size distribution 

 (16-160 fj-m equivalent diameters) were deter- 



'K. Bliss, Oceanographer, Southwest Fisheries Center, 

 La.Jolla, CA, 92038, pers. commun. May 1985. 



"Reference to trade names does not imply endorsement by 

 the National Marine Fisheries Service, NOAA. 



mined with a Coulter model Ta counter with a 

 280 \xm pore configiu'ed to count particles in a 

 20-200 mL sample volume. Counts usually ex- 

 ceeded 40,000 per sample. These determinations 

 stopped part way through the study owing to 

 equipment malfunction at the second site. 



Microplankton 



Microplankton samples were aliquoted prior to 

 counting. A 0.5 mL Stempel pipette was used 

 10-40 times to withdraw a subsample from the 

 well-stirred original sample after adjusting the 

 original volume to 750 niL. When 10 Stempel 

 subsamples yielded too many plankters to enu- 

 merate, the sample was divided with a Folsom 

 splitter to yield a countable fraction from at least 

 10 Stempel subsamplings. 



Microplankton samples were enumerated in 

 covered chambers with a Wild dissection micro- 

 scope (at 250 magnification) to determine mean 

 concentrations of larval anchovy food organisms 

 in the upper 50 m. Food organisms are here 

 assumed to be those having ingestible dimen- 

 sions (20-160 |jLm width), no pronounced spines 

 or processes that would interfere with ingestion 

 or with gut wall integi'ity, and, except for ingest- 

 ible eggs, some degree of motility (Rojas de 

 Mendiola 1974; Arthur 1976). Food concentra- 

 tions and rations given here are underestimated 

 because larvae are known to take organisms 

 smaller than were retained by the meshes of the 

 75 iJtm mesh net used (Rojas de Mendiola 1974; 

 Arthur 1976). The sampled food fraction prob- 

 ably represents the major part of available food 

 rations; although less in number, this fraction 

 is greater in volume than the unfiltered frac- 

 tion and is captured selectively by larvae (Lasker 

 and Zweifel 1978; Theilacker" and Dorsey 1980). 

 The food retained by the net is assumed here to 

 be the majority of, and proportional to, the total 

 rations available to larvae over the 50 m layer 

 sampled. 



Zooplankton 



Zooplankton counts were made on all organ- 

 isms collected in the 333 |xm mesh vertical tows. 

 Samples were not ahquoted. The net tow method 

 used precluded quantitative representation of 

 faster or rarer organisms, including some types 

 of potential predators on fish larvae. Determined 

 for each sample were number of species, number 

 of specimens of each species, and sex ratios 

 (where applicable) for the following major tax- 



675 



