162 



Fishery Bulletin 98(1) 



3 



-o 





Females 



«=255 



10 20 30 40 



50 10 20 30 40 50 



Fin spine age (years) 



Figure 7 



lA) Predictions of length at age from von Bertalanffy models for male and female 

 Atlantic sturgeon (black curves) fitted observed length at age (grey squares) for 

 females better. (B) Plots of residuals from the growth models reveal a difference 

 between fish of size-at-entry (152 ±5 cm TL; black squares); and fish of TL above 

 or below size-at-entry (gray squares). Note different y-axis scales for males and for 

 females. Ages derived from fin-spine sections. 



Growth is difficult to model accurately in long- 

 lived species (Mulligan et al., 1987). Because little 

 bias was detected in our readings (aging precision 

 was unaffected by age), discrepancies in age esti- 

 mates should not introduce systematic error in esti- 

 mates of population parameters such as mortality 

 rate, K, and L^. High variability in observed length 

 at age indicates that length may be a poor predictor 

 of age for Atlantic sturgeon. This high variability 

 may result from amplification of early growth dif- 

 ferences over a long life span. Variance in size at 

 age may also have resulted from the nature of the 



Hudson River Atlantic sturgeon fishery. The imposed 

 minimum size limit on Atlantic sturgeon commercial 

 fisheries (>152 cm TL) may have biased the sample 

 used for growth estimates and could have caused 

 K and L^ to be over- and under-estimated, respec- 

 tively (Fig. 8). The von Bertalanffy curve may have 

 been driven upwards at younger ages by the poten- 

 tial maximum growth rate of the population. For 

 instance, as soon as the fastest growing members of 

 an age class exceeded the minimum size limit, they 

 were harvested. At later ages, the curve is pulled 

 downward as slower growing individuals enter the 



