FISHERY BULLETIN: VOL. 87, NO. 3, 1989 



On 13 June (age 2 days) 24,840 larvae were 

 released into a large artificial outdoor basin — 

 2,000 m^ in volume, 600 m" in surface area, and 

 3.5 m at maximum depth. The basin had been 

 filled with seawater pumped from a depth of 19 

 m. At the time the larvae were introduced, 

 phytoplankton and zooplankton production was 

 high. The basin was drained on 12 August, and 

 the remaining herring larvae in the basin were 

 collected (age = 62 days). 



The larvae in the basin were sampled daily 

 using a two-chambered plankton net of 500 \i.m 

 mesh and a total sampling area of 0.3 m^. The net 

 was drawn diagonally across the basin at a depth 

 of 2 m and the total volume sampled was 7.5 m^. 

 All the sampled larvae were preserved in 80% 

 buffered ethanol. A more detailed description of 

 the basin experiment is given in Wespestad and 

 Moksness (1989)^ The standard length (snout to 

 the tip of the notochord or hyplural plate) of the 

 larvae/juveniles were measured to the nearest 

 1.0 mm. The largest otoliths, the sagittae, were 

 removed and mounted on a glass plate with clear 

 nail polish. The dry weight of each individual 

 was measured to the nearest ± 1 (jig, after drying 

 at 60°C for 24 hours. Otoliths of herring juve- 

 niles over 30 mm had to be ground, to expose 

 growth rings. This was done with fine gi'it paper 

 (30 |jLm and then 0.3 |xm). The maximum magnifi- 

 cation that could be used to read the gi'owth 

 rings in the microscope was x 400, owing to 

 insufficient light penetrating the section. Table 1 

 gives an overview of the number of larvae used 

 in otolith analyses. A detailed description of the 

 otolith analyzing system and the method used 

 are given in Andersen and Moksness (1988). 



RESULTS 



The relationship between the estimated age 

 (estimated number of rings) and the actual age of 

 the herring larvae is shown in Figure 1. The 

 residuals are shown in the same figure. The rela- 

 tionship was linear, and the deposition rate was 

 not significantly different from one increment 

 per day from age 8 days of the larvae (/-test; t = 

 0.08, df = 50). The residuals were equally dis- 

 tributed around zero indicating no trend in the 

 data. The discrepancy did not tend to change 

 sign or range with the age of the larvae (Fig. 2), 



Table 1 . — The number of larvae ex- 

 amined for daily increments by date. 



'Wespestad, V.. and E. Moksness. 1989. Observations 

 on the gi-owth and survival during the early life history of 

 Pacific herring, Clupea pallasi, from Bristol Bay, Alaska, in 

 a marine mesocosm. Submitted Fish. Bull., U.S. 



10 



a 

 ■o 



10 



■10 



-20 



B 



— 1 — 



20 



10 20 30 40 50 60 



Actual age in days 



-■ r 



70 



Figure 1. — Relationship between estimated and actual age 

 of Pacific herring in days (A), y = -8.3 + 1.0144 x,r = 0.96; 

 and the pattern of the residuals (B). 



indicating that the frequency of daily incre- 

 ments in the otoliths did not change with the 

 age of the larvae. The standard deviation of 

 estimated age from real age was ±4.2 days with 

 a range from -12 to -Hi 1 days; therefore, there 

 is little correlation between estimated age and 

 length (Fig. 3). Apparently, there is no relation- 

 ship between the rate of otohth ring deposition 

 and larval growth. 



All three relationships between larval stan- 

 dard length and otolith radius exhibited a good 

 fit to the data (Fig. 4, r > 0.96), but the first, the 



510 



