BARNETT ET AL.: DISTRIBUTION OF ICHTHYOPLANKTON OFF SAN ONOFRE. CALIF. 



indicates that this species must be vertically 

 stratified beyond the 45 m contour. This agrees with 

 the findings of Ahlstrom (1959) in which the majority 

 of E. mordax larvae occurred above 50 m. In contrast, 

 Pleuronichthys veHicalis peaked in abundance at 4-5 

 km rather than extending offshore as in the two- 

 dimensional analysis (cf. Table 2 and Fig. 3K). This 

 result may have occurred because the tests used in 

 the two-dimensional analyses failed to distinguish 

 between offshore blocks due to the small number 

 (27) of non-zero observations for this species. 



Four of the five taxa lacking statistically significant 

 two-dimensional patterns (Chromis punctipinnis, 

 Paralabrax spp., Parophrys vetulus, Peprilus siml- 

 imus) appeared to be most abundant beyond 4-5 km 

 when considered in one dimension (Table 2). The 

 fifth, Pleuronichthys ritteri, peaked in abundance at 

 4-5 km from shore. 



Ontogenetic Pattern Changes 



Larvae of the three most abundant species were 

 divided into size groups, which were analyzed 

 separately for spatial pattern. To prevent temporal 

 bias in the patterns, only 1978 data were used since 

 they covered a full year. Larvae of two sciaenids, 

 Genyonemus lineatus and Seriphus politus, were each 

 divided into groups corresponding to developmental 

 stages. Preflexion larvae, with straight notochords 

 and no hypural development, were analyzed 

 separately from more fully developed, and pre- 

 sumably more mobile, flexion and postflexion larvae. 

 Hypural development was found to begin at 3.8 mm 

 for G. lineatus and at 4.1 mm for S. politus. Similarly, 

 Engroulis mordax larvae were divided into early and 

 late developmental stages, but this was done on the 

 basis of size alone and did not correspond to flexion 

 of the notochord. Early preflexion larvae (<6 mm), 

 termed "early stage", were analyzed separately from 



other larvae, termed "late stage". One hundred lar- 

 vae or all specimens, whichever was less, were mea- 

 sured for each species in each collection. When only 

 the first LOO larvae were measured, the proportions 

 of the various size classes were applied to the total. 



To examine the ratio of older to younger larvae, the 

 total number in each sampling block (Fig. 1) was 

 calculated, using a longshore dimension of 1 m, i.e., 

 number in block, 



N b = N  L, 



where N is number under 1 00 m 2 of sea surface in the 

 block, and L is the onshore-offshore extent of the 

 block in hundreds of meters. 



The patterns of all three species were more 

 nearshore and epibenthic for older larvae (Table 3, 

 Fig. 5). The ratio of older to younger larvae was about 

 1:2 for all three species (transect totals, Table 4). 

 This ratio increased in the shoreward blocks for G. 

 lineatus and 8. politus, reaching maxima in blocks A 

 and B. The ratio of older to younger E. mordax larvae 

 was maximum in blocks C and D. The remarkable 

 aspect of the E. mordax data is that there were far too 

 few eggs in the nearshore zone to account for the 

 numbers of larvae. The ratio of total E. mordax lar- 

 vae to eggs was about 28:1. The median size of the 

 larvae was about 6 mm, corresponding to an average 

 age of roughly 10 d (Methot and Kramer 1979). 

 Zweifel and Lasker (1976) found a time to hatching of 

 2.5 d (at about 16°C). The ratio of 10-d-old larvae to 

 eggs thus has an upper limit of the order 4:1 in the 

 absence of mortality, implying at least a sevenfold 

 excess of larvae in these nearshore samples. The 

 minimum diameter of E. mordax eggs during the 

 months of maximum egg abundance is about twice 

 the mesh opening of the plankton nets used, so that 

 sampling deficiencies for these immobile objects 

 should be negligible. 



Table 3.— Geometric mean abundance (no. 100m 3 ) with 95% confidence bounds (C.B.) for younger and older age 

 groups of larvae of EngrauHsm&rdax, Genyonemus lineatus, and Seripkus politus, showing statistically signifi- 

 cant cross-shelf patterns off San Onofre, Calif. Groups of strata which differ significantly in mean abundance 

 are ranked from highest to lowest. Refer to Figure 5 for locations of these groups. 



105 



