Hernandez and Castro: Lar^/al growth of Engiaulis nngens 



705 



mination of growth rates was still missing, we first 

 determined the age of deposition of the first ring 

 increment on larval otoliths and then described the 

 growth functions of larvae caught in the wild, using 

 three common growth models (the linear, Gompertz, 

 and von Bertalanffy). Finally, we determined the 

 mean growth rates for larval anchoveta spawned 

 over the entire sampling winter season, and for two 

 groups of cohorts spawned during that season but 

 which faced different environmental conditions. 



Methods 



73.5 



73.0 



36.5 



Larval growth in the wild 



Larval anchoveta were collected in eight cruises from 

 a grid of nine stations in the coastal zone off Talca- 

 huano, central Chile (Fig. 1), during the winter of 

 1995 (12, 18 July; 1, 8, 17, 30 August; 4, 11 Septem- 

 ber). In each cruise, ichthyoplankton samples were 

 collected with a bongo net (mesh size: 500 pm, dia- 

 meter of mouth of net; 60-cm) equipped with a flow 

 meter to quantify the volume of water sampled from 

 the surface to a depth of 40 m. Once on board, half 

 of the samples were preserved in 4% formalin and 

 the other half in 96^^ ethanol for otolith analysis. At 

 all stations, sea water samples were collected from 

 nine depths (0, 5, 10, 15, 20, 30, 40, 60, 80 m) in 

 4-liter Niskin bottles for determination of tempera- 

 ture and salinity, and for identification and quan- 

 tification of microplankton (dinoflagellates, copepod 

 eggs, and copepod lai-vae) as potential larval food. 



In the laboratory, anchovy larvae from both sub- 

 samples were identified, sorted, and counted. From 

 the subsamples preserved in ethanol, 112 lai-vae 

 within a size range between 5.68 and 20.74 mm (cor- 

 rected for shrinkage, see below) were measured and their 

 otolits were extracted and mounted in immersion oil. Oto- 

 lith ring counts and otolith diameters were determined 

 with the aid of a light microscope attached to a video 

 camera and monitor to facilitate the reading of daily rings. 

 Each otolith was counted twice and those counts where the 

 readings differed in three or more rings were discarded. 

 Lai^val lengths were corrected for shrinkage by using the 

 algorithms proposed by Theilacker ( 1980 ) for lai-val north- 

 ern anchovy (Engraulis inordax). Three models were used 

 to describe the growth and growth rates of larval anchovy: 

 the linear, Gompertz, and von Bertalanffy models. These 

 models were used to describe growth for 1 ) all larvae 

 spawned during winter (1 July-11 September) of 1995, 

 and 2) for two groups of cohorts spawned during a) 1 July 

 through 17 August, and b) from 18 August through 11 Sep- 

 tember To classify the larvae as belonging to the first or 

 second period, spawning dates were backcalculated as "the 

 number of rings -i- 2" (see day of first ring deposition in the 

 "Results" section). All statistical tests (regressions, analy- 

 ses of variance, covariance, and models) were carried out 

 with the commercial statistical software package STATIS- 

 TICA, 1993. 



37.0 



36.5 



37.0 



73.5 



73 



w 



Figure 1 



Survey area of ichthyoplanktun during the 1995 anchoveta 

 winter spawning season ofT Central Chile. Dots represent 

 sampling stations. 



First increment deposition 



Anchovy eggs were collected from a coastal station off Tal- 

 cahuano during the winter spawning season on 1997. At 

 that station, gentle vertical tows were carried out with 

 a bongo net from to 40 m for collections of ichthyo- 

 plankton. The samples were placed in a 20-gal cooler at 

 12-13°C and transported to the Universidad de Concep- 

 cion Marine Station at Dichato. The transit time to the 

 station took about 45 min. At the laboratory, between 

 20 and 60 anchovy eggs were sorted and placed in ten, 

 700-mL transparent plastic jars and incubated at 10°C 

 and 14°C in light and dark cycles of 14:10 h and 12:12 h, 

 respectively. One third of the water contained in the jars 

 was replaced daily throughout the duration of the study. 

 All larvae hatched in a given day were transplanted to 

 new jars containing water at the same temperature as 

 that used for hatchmg and in which Isocrysis sp. and pow- 

 dered lanal food were added daily. Every day a variable 

 number of larvae were extracted from the jars for determi- 

 nation of fresh larval length, preserved in ethanol, remea- 

 sured. and dissected for otolith analysis (ring counts and 

 diameter). 



