SKILLMAN and YONG: GROWTH CURVES FOR TWO MARLINS 



Table 2.-Striped inarlin von Bertalanffy growth parameters hy 

 cohort and sex. The parameter estimates, L^ (asymptotic 

 maximum fork length), K (rate of proportional growth), and („ 

 (theoretical time at which the tish would have zero length), are 

 given for model 1 (upper row) and mode! 2 (lower row). 



Results— Analysis of Pooled Data 



In the preceding section, it was shown that there 

 were no demonstrable differences in the growth 

 parameters between different cohorts; hence, the 

 estimates for L^, K, and /q could be averaged to 

 provide a pooled estimate. In addition, it was not 

 possible to consistently show significant differ- 

 ences between quarters or years in the number of 

 age-groups separated (S= 11.78, P>0.05), the mean 

 lengths of age-groups (5= 15.51, PsO.05, for males 

 in third quarter, all seven remaining 5 values had 

 P>0.05), or the percent representation of age- 

 groups (S= 57.18, PsO.Ol, for males in the second 

 quarter, S = 49.57, P^O.Ol, for females in the third 

 quarter, all six remaining S values with P>0.05). 

 Neither was it possible to show a trend in mean 

 lengths among cohorts using a series of sample 

 runs tests (one test out of 18 deviated from 

 random at the 0.05 level). We interpreted these 

 results to mean that the yearly samples were 

 homogeneous and that at least approximately a 

 steady state existed. Therefore, the yearly 

 frequency data were pooled; and ENORMSEP was 

 used to separate age-groups. 



The mean lengths and percent representation of 

 age-groups by quarter (Table 3) were quite similar 

 to the values found using the yearly data (Table 1). 

 Quantitatively evaluating the goodness of fit, the 

 chi-square values were found again to be rather 

 large. Qualitatively, however, the shape of the 

 frequency distributions was consistent between 

 sexes within quarters and generally so between 

 quarters within each sex (Figure 2). The plots for 

 the third quarter, the off-season for striped marlin 

 in Hawaii, did not show much of a pattern at all. 

 Also, the shapes of the plots for the pooled data 

 analysis were similar to those for the analysis of 

 individual cohorts (not shown). 



Mean lengths for females and males exhibited a 

 fairly smooth progression (Figure 3). As in the 

 analysis of cohorts, there was an age-group with 

 mean length of about 167 cm in the third quarter 

 that did not fit into the progression of age-groups. 

 Again, it was assumed that this age-group 

 belonged to a different spawning stock and should 

 not be used in the calculation of the growth curve. 

 For females, there were 11 age-groups in the 

 progression whereas there were 10 or 12 but never 

 11 for the analysis of cohorts. For males, there 

 were 12 age-groups in the progression whereas 

 there were 9 to 11 in the analysis of cohorts. The 

 smallest fish were recruited into the fishery in the 

 fourth quarter and progressed through the fishery 

 until the largest fish passed from it in the third 



Table 3.-Statistics for striped marlin age-groups by quarter an^d 

 sex, pooled over all years. Estimates of mean fork length, FL\ 

 percent representation of the age-group, %; the numerical 

 sample for size of the age-group, n; the total sample size, A^; and 

 the chi-square goodness of fit value, x^, were obtained from the 

 computer program ENORMSEP. 



559 



