FISHERY BULLETIN: VOL. S7. NO. 3, 1989 



Two sample sets, both of which provided com- 

 plete water column coverage, were utilized in 

 this study. A small sample set, used to make a 

 rough approximation of feeding chronology, con- 

 sisted of plankton samples collected twice during 

 the day and twice at night within a 24 h period, 

 on two occasions (25 July and 21 September 

 1978). These samples were collected along the 8 

 m isobath 1 km south of the San Onofre Nuclear 

 Generating Station (SONGS) (July) and along 

 the 13 m isobath just north of SONGS (Septem- 

 ber). A larger sample set, used to examine larval 

 diets, consisted of plankton samples collected at 

 night at approximately monthly intervals be- 

 tween 6 September 1978 and 23 September 1979, 

 along a randomly selected isobath within each of 

 five offshore blocks. These blocks were defined 

 by depth contours: (A) 6-9 m, (B) 9-12 m, (C) 

 12-22 m, (D) 22-45 m, and (E) 45-75 m. The 

 blocks were arrayed between 0.5 and 7.2 km 

 from shore, approximately 1 km south of 

 SONGS. 



SONGS Unit 1, a 500 megawatt power plant, 

 operated throughout the study period. Unit 1 

 has been shown to have had only very localized 

 effects (Marine Review Committee 1979^) and it 

 is unhkely to have influenced the results of this 

 study. 



Three different types of gear were used to 

 collect both sample sets: a Brown Manta net 

 (Brown and Cheng 1981), to sample the neuston 

 (top 16 cm); a Brown-McGowan opening-closing 

 bongo net, to sample the midwater column; and 

 an Auriga net^, to sample the epibenthos (within 

 approximately 67 cm of the bottom). All three 

 types of nets were constructed of 0.333 mm mesh 

 Nitex^, fitted with flowmeters, and towed at ca. 

 1 m/s to sample a target volume of 400 m\ All 

 samples were fixed in 10% seawater-formalin. 



In the laboratory, the plankton samples were 

 subsampled with a folsom plankton splitter and 

 the fish larvae were sorted from the subsamples 

 at (3-10 X magnification under dissecting micro- 

 scopes. Larvae utilized in the feeding studies 



'Marine Review Committee. 1979. Interim report of the 

 Marine Review Committee to the California Coastal Com- 

 mission part 1: General summary of findings, predictions, 

 and recommendations concerning the cooling system of the 

 San Onofre Nuclear Generating Station. Mar. Rev. Comm. 

 Doe. 79-02, p. 1-20. Marine Review Committee of the Cali- 

 fornia Coastal Commission, 631 Howard Street, San Fran- 

 cisco, CA 94105, 



^MBC Applied Environmental Sciences, 947 Newhall 

 Street, Costa Mesa, CA 92627. 



^Reference to trade names does not imply endorsement hy 

 the National Marine Fisheries Service, NOAA. 



were randomly selected from among those 

 sorted from a subsample; a maximum of 100 

 specimens was selected from any subsample. 

 These larvae were measured to the nearest 0.1 

 mm notochord or standard length, separated by 

 developmental stage (preflexion = Pr; notochord 

 flexion plus postflexion = FP), placed in a 

 glycerin- water solution, and dissected with fine 

 insect pins. The gut contents of each specimen 

 were identified to the lowest possible taxon us- 

 ing a dissecting (50 x) or compound ( 1 00^50 x) 

 microscope, as appropriate, and enumerated. The 

 number of specimens dissected is listed by 

 species, survey, and sample in Table 1. 



Feeding incidence (%FI = percentage of 

 larvae examined that contained at least one food 

 item) was calculated with 95% confidence limits 

 for each larval stage of each taxon. The 95% 

 confidence hmits were approximated as ±1.96 

 ipq/n)'^'- + Vzn, where p = the proportion contain- 

 ing at least one food item, q = I - p, and v = the 

 sample size. Percent frequency of occurrence 

 (%F0) and percent of the total number (%N) of 

 prey ingested by each larval fish stage were cal- 

 culated for each prey type. 



RESULTS 

 Feeding Chronology 



Feeding incidence was calculated for Seriphus 

 politus, Paralahrax spp., and Paralichthys cali- 

 fomicus which were collected in the day/night 

 sample sets in order to examine feeding chro- 

 nology (Table 2). Larval Genyonemus lineatus 

 and Atherinopsis califomiensis did not occur in 

 these samples, and too few Leuresthes tenuis 

 were available to warrant examination. 



During the day, 82%. of the S. politus larvae 

 contained at least one food item, while at night, 

 72% contained food. Feeding incidence increased 

 from the morning through the evening, reaching 

 a maximum of 94% for the larvae collected be- 

 tween 2003 and 2101 PST. However, since the 

 feeding incidences were all quite high, and the 

 95% confidence limits about %FI broadly over- 

 lapped for all three morning through evening 

 sampling episodes, it seems likely that there 

 were no real differences in feeding incidence 

 over this period. After midnight, feeding inci- 

 dence was reduced to 33%, and the 95% confi- 

 dence limits did not overlap with either the 

 morning or evening values, indicating that feed- 

 ing incidence indeed was reduced after midnight 

 (Table 2). 



Paralahrax spp. feeding incidence was 68% 



570 



