FISHERY BULLETIN: VOL. 70, NO. 3 



polyethylene bag, which was suspended in a 

 4.6-m diameter pool (13.2 kliter) . The pool also 

 contained about 15 northern anchovy (EngrauUs 

 mordax) larvae. A 1,500-w mercury lamp that 

 was suspended about 1 m above the bag provided 

 illumination and a source of heat. The larvae 

 were fed daily with plankton collected in Mis- 

 sion Bay, Calif., and live Artemia. On the 29th 

 day the larvae were released into the pool. Feed- 

 ing continued with plankton and live Artemia up 

 to the 45th day, after which frozen Artemia and 

 dry trout food were substituted. The daily ration 

 was about 10% of the estimated total weight of 

 fish alive on a given day, based on the average 

 weight of fish sampled in the previous month. 



On the 31st day the density of sardine larvae 

 was noticeably reduced. Since few dead larvae 

 were found during the previous days, the ancho- 

 vies (average length 66 mm) were suspected of 

 feeding on the sardines (average length 24 mm) . 

 Analysis of stomach contents of three anchovies 

 revealed that they were indeed feeding on the 

 sardines.^ 



During the first few months of the experiment, 

 water was added to the pool only periodically 

 to remove surface scum. But beginning on the 

 61st day, seawater was circulated continuously 

 through the pool at an initial rate of about 19 

 liter/min. This rate was increased on the 180th 

 day to about 34 liter/min. 



the third month and continuing through the 12th 

 month, about 20-24 fish/month were sampled. 

 After the 12th month, the sampling rate was re- 

 duced to 11-14 fish/month until an unusually 

 large mortality from unknown causes reduced 

 the population to 12 fish in the 16th month. Two 

 more samples were taken thereafter; in one sam- 

 ple four fish were measured for length and re- 

 turned to the pool. After the 24th month, the 

 seven surviving fish were sampled and the ex- 

 periment was terminated. 



Standard length, total weight, gonad weight, 

 and six scales removed from the body area at 

 the tip of the pectoral fin (see Walford and 

 Mosher, 1943:4) were obtained from most sam- 

 ples. 



Water temperature was recorded daily on a 

 thermograph. The monthly mean temperature 

 ranged from 15.3° to 25.0°C and was generally 

 2.9 °C higher than the monthly mean surface 

 temperature recorded off the Scripps pier. La 

 Jolla (Kimura, 1970), site of the water intake 

 for the experimental pooP (see Lasker and Vly- 

 men, 1969), 



Salinity measurements were made by daily ti- 

 trations from surface water sampled off Scripps 

 pier. The mean values ranged from 33.09 to 

 34.38^r. It was assumed that water salinity was 

 the same off the Scripps pier and in the experi- 

 mental pool. 



SAMPLING 



Fish were sampled and sacrificed at various 

 intervals during most of the duration of the ex- 

 periment. About 7-17 larvae were sampled daily 

 from the polyethylene bag during the first month. 

 During the second month 6-26 larvae were sam- 

 pled primarily on a weekly basis. Starting with 



SCALE MEASUREMENTS 



Scales were placed between two glass slides 

 and viewed on a scale projector that magnified 

 them 30 times. All marks that appeared to be 

 annuli were recorded and the widths from focus 

 to mark and focus to margin in the anterior field 

 were measured. 



^ To our knowledge there have not been previous re- 

 ports on predation of sardine larvae by anchovies. The 

 significance of our observation is that predation by an- 

 chovies may have contributed to the decline of the Pa- 

 cific sardine population. Our anchovies were hatched 

 from eggs and were 44 days older than the sardines. 

 This difference in age (size) between the species is also 

 found in the wild; the northern anchovy has a peak 

 spawning period during January-March and the sardine, 

 during April-June (Ahlstrom, 1966). The young of both 

 species coexist in the California Current; thus it is con- 

 ceivable that predation by anchovies is an important 

 source of sardine natural mortality. 



ANALYSIS 



Gonad weight was expressed as a percentage 

 of total weight. This was designated a gonad 

 index, or relative measure of sexual development. 



The relative "fatness" of a fish was estimated 



' The experimental pool was located at the Southwest 

 Fisheries Center, about 1 km from the Scripps pier. 



1044 



