292 



Fishery Bulletin 98(2) 



local precipitation and runoff from South American 

 rivers, particularly the Orinoco River, affect surface 

 salinity which ranges from 35.57ct in the dry season 

 to less than 27%c in the wet season. However, the 

 salinity of deeper waters is not greatly affected and 

 ranges from about 36.8'^<r in the wet season to 36.59cc 

 in the dry season (Gade^). 



Methods 



Monthly samples of yellowedge and yellowmouth 

 grouper were obtained from commercial fish landings 

 at a fish processing plant on the island of Tobago. 

 Fish were caught by Antillean fish traps (Munro et 

 al., 1971) on the continental shelf and slope to the 

 northeast of Trinidad and northwest of Tobago in 

 depths of 37-128 m (Fig. 1). A total of 729 yellowedge 

 and 116 yellowmouth groupers were obtained during 

 the study period. 



Total length (mm) was recorded for the yellowedge 

 grouper, whereas fork length (FL) was recorded for 

 the yellowmouth grouper because, in most cases, the 

 filamentous rays of the caudal fin of this species 

 were damaged. Total body weight (g) was recorded 

 for both species. For most fish the left sagittal otolith 

 was removed unless broken or lost, in which case 

 the right was obtained. Otoliths were stored dry in 

 labelled envelopes. 



For sectioning, otoliths were embedded in Spurr 

 resin (Spurr, 1969) and allowed to harden overnight. 

 One or two 0.5-mm transverse sections of each oto- 

 lith were taken through the focus along a dorsoven- 

 tral plane with a high-speed circular saw. Sections 

 were ground and polished with several grades of sili- 

 con carbide paper. They were placed in glycerol and 

 viewed against a black background with reflected 

 light under a dissecting microscope at a magnifica- 

 tion of 20x. Alternating opaque and translucent bands 

 were visible and the former were counted as annuli. 



For each species, otolith radius (distance from the 

 core to the otolith edge) and the distance from the 

 core to the distal edge of each opaque ring were mea- 

 sured under a binocular microscope with an ocular 

 micrometer ( 1 micrometer unit=0.125 mm). All mea- 

 surements were taken along the ventral surface of 

 the sulcus acousticus. In order to validate that rings 

 were formed annually in the yellowedge grouper, the 

 monthly mean marginal increment ratio, that is, the 

 marginal increment (distance between the distal edge 

 of the outermost annulus and the otolith margin) 

 divided by the distance between the distal edge of the 

 penultimate annulus and proximal edge of the outer- 

 most annulus (Bullock ot al., 1992), was calculated. 

 Because of difficulty in measuring annulus radius 



in the yellowmouth grouper, it was not possible to 

 carry out this analysis in sufficiently large monthly 

 samples. Instead, for this species the monthly fre- 

 quency of occurrence of otoliths with an opaque edge 

 was determined. Mean monthly marginal increment 

 ratios in the yellowedge gi'ouper and frequency of 

 occurrence of opaque edges in the yellowmouth grou- 

 per were compared by using the chi-square goodness- 

 of-fit test for circular data (Zar, 1974). Least squares 

 linear regression of otolith radius on fish length was 

 carried out and the relationship used to backcalcu- 

 late lengths offish at earlier ages. 



The von Bertalanffy growth function was used to 

 describe growth (Ricker, 1975): 



L,=L„(l- 



-Kit-t,. I 



where L^ = total or fork length (mm) at time t 

 (years); 

 L^ = the asymptotic length; 

 K = the growth coefficient; and 

 <ii = theoretical age at zero length. 



This function was fitted to observed lengths at age by 

 using the FISHPARM program (Prager et al., 1987). 

 The length-weight relationship was determined by 

 least squares linear regression with logarithmically 

 transformed data: 



W = aL^ 



where W = body weight (g); 



L = fish length (mm); and 

 a and b are constants. 



Results 



Validation 



Satisfactory annuli counts were made on 326 of the 

 367 yellowedge grouper otoliths examined. A ran- 

 domly chosen subsample of 63 otoliths was read by 

 an independent reader and a 95'^ agreement was 

 found. This species showed a significant difference 

 among monthly mean marginal increment ratios 

 {P<0.001). with elevated values (greater than 50'^) 

 occurring from March to September (Fig. 2). This 

 finding suggests that there is one main period of 

 annulus formation during the year from October to 

 February, even though otoliths with opaque edges 

 were observed throughout the year. 



Growth of the otolith was proportional to growth 

 in length of the fish and the relationship between 

 total length (TL) and otolith radius (OR) was 



