FISHERY BULLETIN: VOL. 78, NO. 1 



ages 2.5, 3.5, 4.5, and 5.5 d. The percentage survi- 

 val was determined on the eighth day. Larvae 

 were fed rotifers (50/ml) on the first day of feeding, 

 and rotifers and copepods (Tisbe sp.) thereafter. 



VulnerabiHty of newly metamorphosed Pacific 

 mackerel (mean length 16.7 mm SL (standard 

 length)) to starvation was tested at 18.9°±0.2° C 

 by transferring fish from the rearing container to 

 containers without food. One group of fish was 

 starved for 4 d, another for 5 d, and a third was fed 

 Artemia salina nauplii. The two starved groups 

 were fed at the end of the starvation period. 



Tail beat frequency and amplitude of larval 

 Pacific mackerel (3-5 mm SL) were determined for 

 routine swimming and burst speeds by analysis of 

 cine films taken at 100 frames/s using techniques 

 described by Hunter (1972). Routine swimming 

 speeds of larval mackerel, at 19° C, were measured 

 by counting the number of squares crossed by a 

 larva (3.7-6.6 mm SL) as it swam over a 1 cm grid 

 on the bottom of the rearing tank for 9-153 s; a 3 

 cm grid was used for larger larvae (7.9-13.1 mm 

 SL) with shorter observation times (3-46 s). The 

 mean of 15-25 visual observations was used as an 

 estimate of speed for the mean length of the larvae 

 in the tank on the day of observation; speeds were 

 adjusted for parallax caused by the difference in 

 depth between the grid and the fish. Speeds of 

 juveniles (>19 mm SL) were measured by timing 

 the fish as they swam a measured distance (35-201 

 cm) around the perimeter of the rearing tank; in 

 this case mean speeds were for individual fish and 

 observation times ranged from 3 to 26 s. 



The size at which Pacific mackerel larvae were 

 capable of ingesting various prey was evaluated 

 by placing them in a 110 1 container with the prey 

 and estimating the number that fed by examina- 

 tion of stomach contents. The type of prey, mean 

 prey size, prey density, and duration of feeding 

 respectively were: yolk-sac anchovy larvae, 2.7 

 mm SL, 8/1, 2 h; A. salina nauplii, 0.2 mm wide, 

 11/1, 4 h; and anchovy eggs, 0.67 mm wide, 10/1, 2 

 h. The mouth width, prey width, and standard 

 length of the larvae were measured and percent- 

 age feeding success was estimated for size classes 

 of larval length and mouth width. The number of 

 fish per size class was >9. Size thresholds for 50% 

 feeding success and 95% success were estimated 

 by probit analysis (Finney 1952) and expressed as 

 a function of mean larval length, or mean prey 

 width/mean mouth width. 



The sizes of food items eaten by Pacific mackerel 

 larvae in the sea was determined by examination 



of the stomachs of 86 larvae taken in routine 

 ichthyoplankton surveys along the California 

 coast. We recorded the length of each larva and the 

 number and maximum width of all identifiable 

 food items (Arthur 1976; Shirota 1970). 



Food requirements were estimated by feeding 

 the rotifer Brachionus plicatilis to 3-5 d old Pacific 

 mackerel larvae. Seven to eight samples of 9-16 

 larvae each were taken over each of three 12-h 

 feeding days, the number of rotifers in the guts of 

 each larva were counted, and the counts converted 

 to equivalent dry weight using the conversion 

 factor of 0.16 /^g/rotifer (Theilacker and McMaster 

 1971). Daily changes in larval weight were esti- 

 mated from mean standard lengths using a 

 length-dry weight conversion given in the results. 

 The rate of gastric evacuation for 4 mm SL larvae 

 was measured. They were allowed to feed for 4 h 

 and then transferred to a tank without food; sam- 

 ples of 13-16 larvae were taken at about hourly 

 intervals until the stomachs were empty. The 

 number of rotifers in stomachs were counted and 

 converted to dry weight, and the rate of gastric 

 evacuation was estimated in terms of dry weight. 

 The daily ration was estimated from the mean 

 stomach contents and the rate of gastric evacua- 

 tion. Gross growth efficiency was estimated in 

 terms of dry weight from the daily ration and 

 weight gain over 24 h. 



Metabolic requirements of Pacific mackerel lar- 

 vae were estimated using a Warburg respirometer 

 and standard manometric techniques (Umbreit et 

 al. 1964) to measure oxygen consumption. One or 

 more Pacific mackerel larvae were added to an 18 

 ml Warburg flask filled with 4.4-8.7 ml of filtered 

 seawater (salinity 33.58-33.93%o). Larvae >0.06 

 mg dry weight were tested individually. Twenty- 

 one tests were made at 18.0°C of larvae or groups 

 of larvae ranging in length from 3.7 to 17.9 mm SL 

 (0.038-12.74 mg) and 14 at 22.0° C, of larvae 3.2- 

 10.5 mm SL (0.025-2.86 mg). Flasks were shaken 

 at 102 times/min for 5 out of every 30 min; read- 

 ings were taken after the first 2 h and continued 

 for 150-360 min. At the end of a test, larvae in each 

 flask were measured, rinsed in distilled water, 

 oven dried to a constant weight, and weighed. 

 Mean weight was obtained for fish tested in 

 groups. All runs were made under normal room 

 illumination, about 700 Ix. Logj^ oxygen con- 

 sumption in microliters Og per hour was regressed 

 on logio body weight for the 18.0° and 22.0° C 

 experiments. As the slopes were close to unity, 

 oxygen consumption was expressed in microli- 



90 



