FISHERY BULLETIN: VOL. 82, NO. 1 



FIGURE 2. — Diagrammatic profile of the 

 study transect showing the 15 strata sam- 

 pled off San Onofre, Calif. Neustonic and 

 epibenthic layers are vertically ex- 

 aggerated. 



Distance from Shore (km) 

 2 3 4 5 6 7 8 

 _1 I I I I 1 I 







- 10 

 20 

 30 o 



ft) 



40 ? 



- 50 3 



- 60 

 70 

 80 



avoid the San Onofre kelp bed, some of the tows in 

 the B and C blocks were offset by about 1 km. 

 VVilcoxon signed rank tests of samples taken from B 

 block and B offset (Fig. 1) showed no significant dif- 

 ferences in species abundances (P > 0.05) between 

 the main block and the offset which could not be 

 related to the inshore-offshore patterns discussed 

 below. 



The transect was sampled monthly in January and 

 February 1978, fortnightly from March through 

 August 1978, and again monthly through September 

 1979. During each of these 28 sampling periods, the 

 five blocks were surveyed once each night for 1-3 

 nights, giving a total of 57 sampling dates for the 21- 

 mo study. 



As noted above, we chose a standard sampled 

 volume of 400 m 3 based on the preliminary study. 

 This volume was large enough to assure a representa- 

 tion of all abundant species throughout the year. 

 Volume was used as the sampling unit, although an 

 argument based on the scale of patchiness could be 

 made for length of tow (i.e., 400 m in each water layer) 

 as the criterion, rather than volume filtered (P. 

 Smith 10 ). Most tows were at least 400 m long. 



Laboratory Procedures 



Samples were sorted for fish eggs and larvae under 

 dissecting microscopes at 10X magnification. The 

 choice of 400 m 3 as the sampled volume was made at 

 a time of year when ichthyoplankton abundance was 

 low (Walker et al. 11 ); consequently the samples from 

 other times of year were larger than necessary to rep- 



,0 P. E. Smith, La Jolla Laboratory, Southwest Fisheries Center, 

 National Marine Fisheries Service, NOAA, La Jolla, CA 92038, 

 pers. commun. 1979. 



"Walker, H. J., A. M. Barnett, and P. D. Sertic. 1980. Seasonal 

 patterns and abundance of larval fishes in the nearshore Southern 

 California Bight off San Onofre, California. Marine Ecological Con- 

 sultants of Southern California, 53 1 Encinitas Boulevard, Encinitas, 

 CA 92024. 



resent the nearshore assemblage. Samples with large 

 plankton volumes were subsampled, using a Folsom 

 plankton splitter before sorting. The size of the sub- 

 sample was set to include at least 100 non-engraulid 

 larvae (the mean number of larvae counted per sub- 

 sample was 277, of which 62.8% was is. mordax). This 

 fraction was usually one-fourth and was seldom 

 smaller than one-eighth. Eggs were sorted from 1%, 

 5%, or 10 f /r (to get at least 100 eggs) of the residue of 

 the fraction sorted for larvae. Sorting efficiency was 

 maintained above 90%. 



Some epibenthic samples contained so much sand 

 and detritus that it was necessary to clean them 

 before sorting, using a flotation technique adapted 

 from Ladell (1936). After removal of large fish and 

 debris, such a sample was mixed with a 40% MgS0 4 

 solution (specific gravity = 1.2) in a large separator 

 fashioned from a 19 1 (5-gal) plastic carboy with the 

 bottom cut off and the neck fitted with a rubber hose 

 and ball valve. Most detritus sank, while plankton 

 floated to the top. The heavy material was drained off 

 and processed once or twice more to ensure separa- 

 tion of the plankton. Checks of the heavy residue of 

 three such samples showed that more than 99% of 

 the larvae were separated by flotation. 



All larvae were identified to the lowest taxonomic 

 category currently possible. Eggs were identified as 

 Engraulis mordax or "other". In some larval cate- 

 gories (e.g., Atherinidae, Goby Type A), our ability to 

 discriminate among species or larval types (sensu 

 Richardson and Pearcy 1977) improved as the study 

 progressed. However, not all of the old collections 

 were reprocessed. When mixed taxa showed 

 seasonal and spatial coherence, they were retained 

 for the analyses presented here. 



Pattern Analysis 



All counts of eggs and larvae were standardized to 

 number/400m-\ Thus the standardized numbers 



100 



