Rivera and Appeldoorn; Age and growth of Coryphaena hippurus 



347 



weight, and sex were recorded. All fish were measured 

 to the nearest millimeter and weighed to the nearest 

 50 grams. Heads were separated from the body at the 

 site of collection and kept frozen. 



The removal and preservation of otoliths followed 

 the methods of Panella (1980) and Brothers (1987). 

 A sagittal (longitudinal) cut through the midline of 

 the frozen head was made with an electric band 

 saw. Under a dissection microscope, sagittae were 

 removed from the sacculi and extraneous tissue was 

 removed. Sagittae were used owing to their rela- 

 tively larger size in comparison with the lapillus and 

 asteriscus. Each pair of sagittae was stored in gly- 

 cerin for clearing in labeled vials. 



Examination of external otolith micrcstructure 

 was used to determine age (Panella, 1971). Sagittae 

 were placed on glass slides in glycerin. Otolith struc- 

 ture was examined and the radius measured under a 

 compound light microscope with transmitted light at 

 200x. Translucent and opaque lines were counted fol- 

 lowing the procedure of Oxenford and Hunte ( 1983). 

 Otolith rings were assumed to be daily lines (Uchi- 

 yama et al., 1986) and were counted from the focus 

 to the edge of the posterior rostrum. 



All otoliths were read twice at random. If there 

 was any difference in reading (10% or more), the 

 otolith was discarded. A subsample of ten otoliths 

 was sent to Hazel Oxenford (Bellairs Research Insti- 

 tute of McGill University, in Barbados) for additional 

 reading and verification of counts. Otoliths were sent 

 in coded vials with no information about length, 

 weight, or sex. Fish length, from which otoliths were 

 sent, ranged from 630 mm FL to 1325 mm FL. 



The relation between fish fork length and daily 

 increment of the sagittae was determined with a pre- 

 dictive linear regression of length on number of rings 

 (Ricker, 1975, Francis, 1990). Following Oxenford 

 and Hunte (1983) and Bentivogho (1988), growth 

 rates were calculated from the linear regressions and 

 reported as mm/day. Analyses were done by coast 

 (north and south) and by sex (male and female). Dif- 

 ferences between growth-rate estimates were com- 

 pared by using a homogeneity of slopes test ( ANOVA) 

 (SokalandRohlf, 1981). 



For a more realistic representation of growth, age- 

 length data were also modeled by using the von Ber- 

 talanffy growth function. 



l,=L^(l-e-'"'-'"'], 



where /, - length (mm) at time t (years); 

 L^- asymptotic length; 

 k = the growth coefficient; and 

 Iq - the hypothetical age at which length 

 equals zero. 



Parameters were estimated by a nonlinear regres- 

 sion with SYSTAT (Wilkinson, 1987). 



Results 



A total of 170 dolphinfish were collected during the 

 eight-month sampling period. From this total, 80 

 were captured off the north coast and 90 were cap- 

 tured off the south coast. The size range from the 

 south coast was broader than that from the north: 

 north = 475 mm FL ( 1.25 kg) to 1283 mm FL ( 18.50 

 kg); south = 381 mm FL (0.70 kg) to 1479 mm FL 

 (25.00 kg); in addition, the largest fish were found on 

 the south coast. From the total sample, 55 were male 

 and 115 were female. Males were slightly larger than 

 females: males = 630 mm FL (2.50 kg) to 1479 mm 

 FL (25.00 kg); females = 381 mm FL (0.07 kg) to 

 1283 mm FL( 19.75 kg). 



The relation between standard length (SL) and fork 

 length (FL) was linear and expressed by the equation 



SL = -1.37 -I- 0.92 FL 



(r=0.99). 



The relation between the logarithms of fish weight 

 ( W) and fork length was linear and expressed by the 

 equation 



log W = -4.42 H- 2.78 log FL 



(r=0.98). 



Otoliths were collected from 22 males and 38 

 females from the north coast, and from 21 males and 

 40 females from the south coast. Otoliths were not 

 collected from all fish owing to difficulties in making 

 the longitudinal cut through the head, to breakage 

 during dissection, or to loss during extraction from 

 the cranial tissue. Thus, for the age and growth 

 determinations, fork lengths ranged on the north 

 coast from 746 mm FL to 1283 mm FL for males 

 and from 475 mm FL to 1222 mm FL for females. 

 Males on the south coast ranged from 625 mm FL to 

 1325 mm FL, and females ranged from 550 mm FL 

 to 1275 mm FL. 



Examination of the external structure of the oto- 

 lith showed clear growth increments. On large oto- 

 liths, increments tended to be tightly spaced on the 

 edge of the rostrum. Reading daily increments in this 

 area was difficult owing to poor resolution even after 

 clearing in glycerin; however, independent readings 

 by Oxenford were within 10%. The oldest individual 

 had 336 increments. 



Figure 2 shows the length and number of otolith 

 increments for all fish. Linear growth rate was 2.52 

 mm FL/day for all fish (Table 1, Fig. 2). Differences in 

 linear growth rates by sex or coast (Table 1) were not 



