892 



Fishery Bulletin 97(4), 1999 



al., 1982). Scales were collected above the lateral line, 

 at the level of the pectoral fin, cleaned with 5% KOH, 

 and mounted on slides. Interpretation of scale rings 

 was carried out with a projector at lOx magnifica- 

 tion by using the method described by Beardsley 

 (1967). The scarcity of adult fish prevented the use 

 of age validation methods; therefore the annual na- 

 ture of the rings was established from validated ages 

 from other areas (Beardsley, 1967; Rose and Hassler, 

 1968). Juvenile fish (n=4084) were measured from 

 65 samplings made between August and December 

 of the same years from small-scale fishery catches. 

 Total weight and sex were recorded for subsamples 

 (adults n =68, juveniles n=282) selected to cover the 

 size distribution of both sexes. Sagittal otoliths were 

 removed, cleaned, and stored in distilled water. Un- 

 broken otoliths («=212) were embedded in epoxy 

 resin and polished with a graded series of sandpa- 

 per and, finally, with 0.3-|im alumina paste. To im- 

 prove clarity, otoliths were moistened with immer- 

 sion oil for several hours before reading. 



Otoliths have a complex structure, with convex 

 sides, pronounced rostrum and antirostrum, and a 

 demarcated sulcus that terminates in a V-shaped 

 excisural notch (Oxenford and Hunte, 1983). We se- 

 lected the dorsal side of the otolith to obtain a sec- 

 tion with a complete increment sequence. Otoliths 

 were read under a light microscope coupled to a high- 

 resolution video camera and monitor system. Growth 

 increments were counted from the core to the edge 

 of the pararostrum (Oxenford and Hunte, 1983). In- 

 cremental counts were made by beginning at the first 

 clearly defined mark that encircled the primordium, 

 which defines the outer edge of the nucleus. Growth 

 increments in juvenile fish otoliths were enumerated 

 along a single axis. In adult fish it was necessary, how- 

 ever, to follow a circuitous path to complete a set of 

 counts, following prominent increments laterally until 

 an area with clear increments was found. Each otolith 

 was read independently by two of the authors; results 

 were accepted only if the readings were coincident or 

 their difference was less than 57i . Growth increments 

 were counted at 500x and verified at lOOOx. Microstruc- 

 tural growth increments were studied on fish of 22^3 

 cm FL («=28) with a scanning electron microscope 

 (SEM). To highlight increment zones, the polished sec- 

 tions were etched for 30 seconds with a 0.1 N HCl solu- 

 tion, or for 60-90 seconds with a 0.2 M EDTA solution. 



Von Bertalanffy growth parameters were esti- 

 mated from age-length relationships obtained from 

 scale and otolith readings by using the FISHPARM 

 program (Prageretal., 1987). These parameters were 

 validated by an independent length-based method 

 by using the ELEFAN I procedure included in the 

 ELEFAN software package (Gayanillo et al.M. Be- 



cause L^ and k are inversely correlated, the growth 

 performance index (0 = 21ogL^ log^) was employed 

 to compare growth rates (Munro and Pauly, 1983). 



The temporal relation between ring number and 

 age was determined by using larvae of known age, 

 obtained from eggs spawned in captivity. Larvae 

 hatched on 25 July 1994 and were reared through 

 the juvenile stage up to 38 days, according to the 

 method described by Kraul (1989). Water tempera- 

 tures during the experiment ranged from 26° to 27°C, 

 within the range of water surface temperature found 

 around the Balearic Islands from June to August, 

 when the species spawns (at 24-27°C; Massuti and 

 Morales-Nin, 1995). The presence of early larval 

 stages has also been reported in the study area dur- 

 ing this period (Alemany and Massuti, 1998). 



To establish the timing of first ring deposition and 

 early otolith growth, larvae were sampled daily dur- 

 ing the first 15 days. Afterwards, three samples were 

 taken at intervals of 7-8 days until the juveniles were 

 38 days old. Larvae were preserved in buffered alco- 

 hol. The standard length (SL) of each larva was mea- 

 sured, and sagittal otoliths (n=93) were removed, 

 cleaned, and mounted on slides with epoxy resin. The 

 otoliths were read with an image analysis system 

 connected to a microscope. An iterative semiauto- 

 matic system was employed for detection of incre- 

 ments, measurement of otolith radius, and age at- 

 tribution. Increment clarity was improved by using 

 polarized light, and enhanced digitized images were 

 used for interpretation. Each otolith was read at least 

 twice at lOOOx magnification with an oil lens. 



Daily periodicity of increments in reared specimens 

 was validated by linear regression analysis of the 

 relation between counts of otolith increments and 

 days. The intercept and the slope were examined with 

 Students <-test. This method was also used to test 

 the isometry in the allometric indices obtained by 

 applying exponential regression equations to the 

 length-weight relationships. 



Results 



The parameters of the length-weight relationship for 

 females, males, and the whole population, including 

 unsexed specimens, are shown in Table 1. Isometric 

 growth was found in males (Ntest; P>0.05l, but for 

 females and the population as a whole, a negative 

 allometry was found (^test; P<0.01). 



Gayanillo. F.C., M.Soriano, and D.Paulv- 1988. A draft guide 

 to the complete ELEFAN. ICLARM Software Project 2, 70 p. 

 International Center for Living Aquatic Resources Management, 

 Manila, Philippines. 



