SKILLMAN and YONG: GROWTH CURVES FOR TWO MARLINS 



450 



400 



350 



-~ 300 



E 



250 



g 200 

 O 



ISO 



100 



50 



T I I r 



FEMALE (o) 



MALE (O 



qI 1 1 1 L 



-J 1 1 I I I I 1 l_ 



AGE IN YEARS 



Figure 5.-Blue marlin von Bertalanffy growth curves by se.x for 

 the analysis of pooled data. Observed mean lengths for female 

 and male age-groups are given. 



all age-groups, and then using only those with 

 estimated numerical representations greater than 

 two individuals. Further analyses were done using 

 age-groups for females, over the same age span as 

 for males, wnth numerical representation greater 

 than two individuals (Table 6). For males, pa- 

 rameter estimates were similar using all age- 

 groups and those age-groups represented by more 

 than two individuals. The standard error of es- 

 timate was smaller for the latter than it was for 

 the case using all age-groups. For females, again 

 the parameter estimates were similar for the two 

 data sets, and the standard errors of estimates did 

 not change appreciably. The estimates for L^ were 

 nearly doubled those for males. In addition, the 

 estimate of L^ for females, using the same age- 

 groups as for males with age-groups represented 

 by more than two individuals, was nearly 3V2 times 

 that for males. 



Discussion 



The estimates of von Bertalanffy growth pa- 

 rameters for male blue marlin differed little 

 whether all of the age-groups were used or 

 whether the less well-represented age-groups 

 were deleted (Table 6). Because the standard 

 errors of estimates were generally smaller for the 

 reduced data sets, we felt that these fits provided 



Table 6.- Blue marlin von Bertalanffy growth parameters by se.x 

 for the analysis of pooled data. The parameter estimates, L^ 

 (asymptotic maximum fork length), K (rate of proportional 

 growth), and ^,( theoretical time at which the fish would have zero 

 length) are given for model 1 (upper row) and model 2 (lower 

 row). 



better estimates of parameters. Although the 

 standard errors of estimates were larger than 

 desirable, they varied from less than 1% to only 7% 

 of the estimated Lr^. Thus, the von Bertalanffy 

 growth model described the data satisfactorily. 

 The mean length estimates for the poorly repre- 

 sented age-groups, which were the youngest and 

 oldest in our samples, should be viewed as 

 approximate. 



For males, estimates of L^, 368.0 and 298.8 cm 

 for models 1 and 2, respectively, bracketed the 

 commonly accepted asymptotic length of about 

 300 cm. If our assumption of a knife-edge limit of 

 143 kg (approximately 300 cm) for males was 

 incorrect, the progression of age-groups would 

 have been expected to increase in length up to this 

 point without approaching an asymptote. Since an 

 asymptote was found, we felt our assumption was 

 valid. 



For females, the von Bertalanffy growth curves 

 seemed adequate for describing the data, but the 

 estimates of growth parameters were not 

 biologically reasonable. Using the same range of 

 age-groups as used for males, the estimates of Lr^ 

 were around 1,000 cm, confirming the visual im- 

 pression that there was little tendency towards an 

 asymptote over this range of ages. Using all of the 

 data, estimates of L^ were 626.6 and 540.2 cm for 

 models 1 and 2, respectively, or approximately 

 1,729 and 1,060 kg, respectively. While these 

 results suggested that there was some tendency 

 towards an asymptote, which is not visually ap- 

 parent in the data, we do not believe that enough 

 older age-groups were included in the regressions 



565 



