FISHERY HULLKTIN: VOL 8(i, NO 1 



major cold events over the past 20 years occurred 

 in 1964, 1967-68, 1970-71, and 1974-75 (Guillen 

 1983). 



The decline and ultimate collapse of the an- 

 choveta fisheries of Peru began in 1970 and was 

 finalized by the intense 1972-73 El Nino; the 

 northern Chilean stock decline started in 1972 

 and was finalized by 1977. Factors facilitating 

 these declines are generally believed to include 

 overfishing and the devastating effects of the El 

 Nino on anchoveta spawning behavior and in- 

 tensity as well as on subsequent recruitment. 

 Competition and/or predation pressure result- 

 ing from increasing abundances and distributions 

 of sardine and mackerel have also been hy- 

 pothesized (Santander and Flores 1983; Serra 

 1983). 



Because of the great socioeconomic value of the 

 dominant pelagic fish species of the Peru-Chile 

 ecosystem, their population fluctuations have re- 

 ceived a great deal of attention over the past 20 

 years. However, coincidental changes in the com- 

 position, abundance, or spawning intensities of 

 other commercially less important and non- 

 harvested species have not been examined. Infor- 

 mation on the changes of these unfished species 

 in relation to hydrographic conditions and fluctu- 

 ations of the dominant pelagic fish stocks provide 

 additional insight into the ecology of the Hum- 

 boldt Current and may elucidate possible causes 

 for the dramatic changes which occurred during 

 the 1970's. 



In the present work we examine the abundance 

 and composition of total ichthyoplankton assem- 

 blages collected off of northern Chile (lat. 18°- 

 24°S) during 1964-73 and 1983 in relation to am- 

 bient hydrographic conditions. "Normal" cold 

 water as well as warm-water and El Nino events 

 occurred during the 19-yr sampling span. We also 

 examine our results with respect to possible 

 chronological change in environmental condi- 

 tions which led to the 1977 anchoveta fishery col- 

 lapse off northern Chile. Our results may be ap- 

 plicable for interpreting coincidental changes 

 in the Peruvian ecosystem and may also be 

 broadly applicable for studies of similar changes 

 in the other eastern boundary current ecosys- 

 tems. 



METHODS 



Samples were collected during 1964-73 and 

 1983 ichthyoplankton surveys conducted by the 

 Instituto de Fomento Pesquero. The area most 



intensively surveyed was a narrow coastal strip 

 extending between Arica and Antofagasta (lat. 

 18°-24°S, long. 70°-72°W; Fig. 1). This area in- 

 cludes one of two major anchoveta (Engraulis rin- 

 gens ) spawning grounds off Chile and the pri- 

 mary sardine iSardinops sagax) spawning area 

 off Chile prior to 1973 (Fig. 2A, B). All samples 

 used for interannual comparisons were collected 

 during late July-September following peak win- 

 ter anchoveta and sardine spawning periods. Be- 

 tween 21 and 87 samples from the 18°-24°S area 

 were analyzed for each of 11 cruises (Table 1). In 

 one case data from two cruises (August and Sep- 

 tember 1968) were pooled to provide adequate 

 coverage. Sampling was done annually from 1964 

 to 1970 and in 1972 and 1973. There was a 10-yr 

 hiatus before regular sampling was resumed in 

 1983. 



The 1964-73 samples were collected with 

 Hensen nets (0.28 m^ mouth opening; 300 ixm 

 mesh). Prior to 1973 the vertical net hauls were 

 50-0 m; in 1973 haul depth was increased to 100 

 m. The 1983 100-0 m vertical hauls were made 

 with WP2 nets (0.25 m^ mouth opening; 

 UNESCO 1968) of 300 ^JLm mesh. Samples were 

 preserved using buffered 5% formalin solution. 

 Sea surface temperature and salinity data were 

 collected at most sampling stations for all but two 

 winter cruises; these data are lacking for 1970 

 and salinity data are minimal for 1967. 



All fish eggs and larvae were removed from 

 samples, and invertebrate zooplankton biomass 

 was measured. Wet weight displacement volume 

 was measured for 1964-73 samples; in 1983 the 

 Yashnov (1959) technique modified by Robertson 

 (1970) was used. A calculated correction factor 

 of 1.44 (±3.34) was applied to the 1983 biomass 

 values to permit comparison with the earlier 

 data. 



All fish larvae were identified to lowest taxon 

 possible and counted. We herein treat the larvae 

 of six commercially important species (anchoveta 

 [Engraulis ringens], Pacific sardine [Sardinops 

 sagax], jack mackerel [Trachurus murphyi; also 

 known as T. symmetricus in U.S.A.], chub mack- 

 erel [Scomber japonicus]. South Pacific men- 

 haden [Ethmidium maculatum], and hake [Mer- 

 luccius gayi]) separately from the other 35 

 identified taxa. These six species are referred to 

 as the "PL" (larvae of pelagic schooling species). 

 The other larval taxa considered together are the 

 "OL". The PL and OL categories are treated sepa- 

 rately because abundances of the PL (especially of 

 anchoveta and sardine) mask abundance rela- 



