PATTERNS OF LARVAL DRIFT IN SOUTHERN CALIFORNIA 

 MARINE SHORE FISHES INFERRED FROM ALLOZYME DATA 



Robin S. Waples' and Richard H. Rosenblatt^ 



ABSTRACT 



A multispecies analysis of allozyme data for 10 marine shore fishes was undertaken to identify patterns 

 of genetic differentiation resulting from larval drift. Previous studies suggest that allele frequencies in 

 these fishes are sensitive primarily to the effects of migration, rather than to natural selection or historical 

 factors. The following patterns recur in most species: 1) Two northern populations (La Jolla, California, 

 and the California Channel Islands) share a relatively high genetic affinity with all other populations, 

 while the two southern populations (Isla de Guadalupe and Punta Eugenia, Baja California, Mexico) are 

 relatively divergent; 2) the two southern populations apparently exchange genes much more frequently 

 with northern populations than with each other; 3) anomalous results for the ocean whitefish, Caulolatiliis 

 princeps, can be understood on the basis of known patterns of larval distribution in this species. The 

 consistency of these large-scale patterns among species with markedly different life history features 

 and dispersal capabilities suggests that the results obtained here may provide insight into the population 

 structure of other species (invertebrates as well as fish) with pelagic larvae. 



Two characteristics of shallow-water marine organ- 

 isms make the analysis of their population structure 

 interesting and challenging. First, adults of these 

 species are restricted to relatively shallow, inshore 

 waters, so adult populations can be isolated from 

 other populations by expanses of deep water or 

 areas of otherwise unsuitable habitat. On the other 

 hand, many marine species have a pelagic larval 

 stage lasting several weeks or months and thus at 

 least the potential for long-distance transport by 

 ocean currents. Indeed, such long-distance disper- 

 sal events are generally invoked to explain the pres- 

 ence of shallow-water marine organisms on oceanic 

 islands isolated by up to several thousand kilometers 

 from possible sources of propagules. 



However, very little is actually known concern- 

 ing the complex process of larval drift, and several 

 questions remain largely unanswered. For example, 

 by what pathways do larvae traverse oceanic 

 barriers separating different populations? Further- 

 more, do recruits arrive at remote areas on a more- 

 or-less continuous basis, or is long-distance disper- 

 sal the result of rare or unique "sweepstakes" 

 events? The answers to these questions are relevant 

 not only to evolutionary biologists seeking to under- 

 stand the processes of differentiation and specia- 



'Scripps Institution of Oceanography, University of California, 

 San Diego A-008, La Jolla, CA 92093; present address: Northwest 

 and Alaska Fisheries Center, National Marine Fisheries Service, 

 NOAA, 2725 Montlake Boulevard East, Seattle, WA 98112. 



^Scripps Institution of Oceanography, University of California, 

 San Diego A-008, La Jolla, CA 92093. 



tion, but also to those who, in order to formulate 

 management policies for marine fishery resources, 

 must determine the degree to which geographic 

 stocks correspond to independent reproductive 

 units. 



Several approaches have been used to address the 

 problem of pelagic dispersal. In some cases, suffi- 

 cient data regarding oceanic currents are available 

 to construct models capable of predicting patterns 

 of larval distribution if time and place of spawning 

 are known. However, as such models are generally 

 based on long-term mean current patterns, they may 

 be misleading if successful dispersal is actually due 

 to anomalous conditions that occur infrequently. 

 Furthermore, few data are available regarding the 

 inshore currents intimately involved in the initial 

 dispersion (or retention) of larvae spawned in 

 shallow water. Marked drifters (e.g., Schwartzlose 

 1963; Tegner and Butler 1985) can provide biologi- 

 cally relevant data regarding current patterns, but 

 such studies rely on retrieval by the human popula- 

 tion at large and thus provide little information 

 about dispersal to remote (and typically poorly in- 

 habited) localities. Tagging studies, although very 

 resource intensive, can provide valuable, direct in- 

 formation regarding oceanic migrations but are not 

 well-suited to the study of larval drift. 



For the above reasons, indirect methods must 

 often be used to estimate the incidence of gene flow 

 in marine organisms. The electrophoretic analysis 

 of protein polymorphisms is one such approach that 

 has seen extensive use in both terrestrial and aquatic 



Manuscript accepted October 1986. 

 FISHERY BULLETIN: VOL. 85, NO. 1, 1987. 



