Rutherford and Houde: The influence of temperature on growth of Morone saxatilis 



317 



"SC- 



UPPER BAY 

 ESTUARY 



POTOMAC RIVER 

 ESTUARY 



Quantico 



I Morganlown 



Figure 1 



Map of sampling areas and stations on the Potomac River and Upper Chesa- 

 peake Bay, 1987-89. 



keys' were developed separately for the Potomac 

 River and the Upper Bay in each year, except 1988, 

 from regressions of otolith-derived ages on larval 

 lengths. In 1988, otoliths of the few larvae collected 

 in the Upper Bay were unreadable because of poor 

 preservation; consequently the age-length key for 

 1988 Potomac River larvae was also applied to Up- 

 per Bay larval catches. 



Larval growth rates were estimated by using both 

 an "aggregate sample" method and a back-calcula- 

 tion method. In the aggregate sample method, stan- 

 dard lengths and otolith-derived ages of the entire 

 sample of otolith-aged larvae were analyzed for each 



cohort in each of the areas. Growth rates of larvae 

 were derived from exponential regressions of larval 

 lengths on ages. Exponential models also were fit to 

 the mean back-calculated length at age of larvae to 

 estimate back-calculated, cohort-specific growth rates. 

 We backcalculated fish lengths at age by using the 

 "biological intercept" method (Campana, 1990). This 

 method does not assume that larvae from all cohorts 

 or populations have the same body length and otolith- 

 radius relationship; thus it allows reconstruction of 

 individual growth histories, which was desirable for 

 our analysis. This method, and other back-calcula- 

 tion methods that assume a constant proportional- 



