HEDGECOCK ET AL.: GENETIC VARIATION IN PACIFIC SARDINES 



lasted, on average, 80 years and ranged in dura- 

 tion from 20 to 200 years (Soutar and Isaacs 

 1969). One or more of these periods of low abun- 

 dance could have been a severe enough bottle- 

 neck to cause loss of variation, but this hypothe- 

 sis may be difficult to falsify. According to Fitch 

 (1969), fossil remains of Sardinops are absent 

 from samples of California Pliocene and Pleisto- 

 cene sediments, whereas evidence of five other 

 pelagic species (Clupea pallasi, Engraulis 

 Diorda.v, Merluccius productus. Scomber japon- 

 icus, and Trachurus symmetricus) is present in 

 at least some samples. This raises the possibil- 

 ity that Sardinops sagax caerulea may be a 

 recent arrival in the California Current System 

 and that the low variation is attributable to a 

 small number of founders rather than to a sub- 

 sequent bottleneck. 



Other than such historical hypotheses, we can 

 pose no ecological explanation of low genetic var- 

 iation (such as provided for decapod crustaceans 

 by Nelson and Hedgecock 1980, for example); 

 the ecology of the Pacific sardine does not appear 

 to be unique relative to those clupeoids having 

 higher levels of variation (Blaxter and Hunter 

 1982). So we are left at present with no compel- 

 ling hypothesis to explain the observation of low 

 genetic variation in the Pacific sardine. 



Structures of Historical and 

 Contemporary Populations of 

 the Pacific Sardine 



Sprague and Vrooman (1962) and Vrooman 

 (1964) described three genetically distinct sub- 

 populations oi Sardinops sagax caerulea on the 

 basis of significantly different frequencies of a 

 C-positive blood factor (13.6% in samples taken 

 from California waters, 6.0% in samples taken 

 from Baja California, and 16.8% in fish from the 

 Gulf of California). Regrettably, though under- 

 standably, electrophoretic separation of allo- 

 zymes has completely supplanted immunological 

 methods for studying population structure. Data 

 comparable in quantity and quality to historical 

 data on serotype frequencies would be difficult 

 to gather today. A considerable drawback to the 

 immunological method is the requirement for 

 fresh blood, whereas allozymes can be readily 

 obtained from fresh or fresh frozen, muscle or 

 visceral tissues. Moreover, allozyme methods 

 allow a much larger survey of genes than does 

 blood typing; this, in turn, provides for statis- 

 tical analyses of genetic diversity that take into 



account the large component of variance among 

 loci (Nei 1978). 



Our finding of low genetic variation across a 

 widespread samphng of Pacific sardines contra- 

 dicts the hypothesis that there are currently 

 genetically different, geographic subpopulations 

 of Pacific sardine. Combined with the recency of 

 the Pacific sardine's reexpansion into the Cali- 

 fornia Current (Wolf et al. 1987), our observa- 

 tions support the alternative hypothesis that this 

 species comprises a single, homogeneous gene 

 pool. Examination of the distributions and abun- 

 dances of sardine eggs and larvae (Ahlstrom 

 1954, 1957) does suggest the possibility of dis- 

 persal around the tip of Baja California, partic- 

 ularly during cold-water (anti-El Niiio) years. ^ 

 Our data on the sharing of rare alleles by widely 

 separated populations support this conjecture by 

 implying a high rate of gene flow throughout the 

 range of the species (Slatkin 1985). 



The present study does not falsify the sub- 

 population hypothesis for historical sardine 

 populations, but our data show that it is unlikely. 

 The former hypothesis requires that only a 

 single southern subpopulation survived the fish- 

 ery collapse to repopulate the Gulf of California, 

 the Pacific coast of Baja Cahfornia Sur, and more 

 recently, the California Current. Data on the 

 frequency of C-positive blood type in contem- 

 porary sardine populations would be useful. 

 However, morphological and life history data 

 also played an important role in past inferences 

 concerning the structure of historical sardine 

 populations (Radovich 1982). The implications of 

 our data on morphological variation among con- 

 temporary populations are discussed next. 



Morphological and Life History 

 Variation Among Historical and 

 Contemporary Pacific Sardine 

 Populations 



Life history traits, such as the schedules of 

 age-specific growth, mortality, and reproduc- 

 tion, and the covariances among these traits, 

 determine responses by fish populations to ex- 

 ploitation (Gushing 1973; Nelson and Soule 

 1987). Indeed, the historical biology of the Cali- 

 fornia sardine fishery and its demise provides an 

 elegant example of this axiom. An important 



-R. A. Schwartzlose, Centre de Investigaeiones Biologicas 

 de Baja California sur and Scripps Institution of Oceanog- 

 raphy, La Jolla, CA 92093, pars, commun. 1988. 



667 



