THORROLD: LIFE HISTORY PARAMETERS IN HERKLOTSICHTHYS CASTELNAUI 



specimens, 5.6-22.5 mm SL. Laird-Gompertz and 

 von Bertalanffy models yielded good and nearly 

 identical fits to the data (r^ = 0.74 for the Laird- 

 Gompertz curve, r^ = 0.75 for the von Bertalanffy 



curve); therefore, only the Laird-Gompertz growth 

 curve is presented (Table 3; Fig. 4). This relation- 

 ship does, however, appear to underestimate growth 

 at ages less than 10. By contrast, the von Berta- 

 lanffy curve underestimated growth at ages greater 

 than 38. 



Table 2.— Validation of ageing using the tetra- 

 cycline tectinique. Fish were preserved 1 1 days 

 after treatment with tetracycline. Table shows stan- 

 dard length (SL), mean number of increments 

 observed between the fluorescent band and the 

 margin of both sagittae ( + standard error), and the 

 range of increment counts on each of the six fish. 



Table 3.— Laird-Gompertz equation and estimated param- 

 eters describing growth of 378 Herklotsichthys castelnaui 

 larvae. The growth model was fitted using nonlinear least- 

 squares regression. STDERR = asymptotic standard 

 error of parameter estimates; C.L. = approximate con- 

 fidence intervals of parameter estimates. 



25.0 n 



20.0 



^ 15.0 H 

 UJ 



Q 



< 10.0 - 



Q 



< 



h- 

 00 



5.0 - 



0.0 



n = 378 

 r^ = 0.74 



I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 

 10 20 30 40 50 60 



ESTIMATED AGE (days) 



Figure 4.— Relationship between standard length and number of growth increments (increment number) on sagittae for larval Herklot- 

 sichthys castelnaui, together with fitted Laird-Gompertz growth curve. 



79 



