150 



Fishery Bulletin 102(1) 



2 ! 



1 

 o 

 -1 



-2 

 -3 



Cohort A 

 Log e weight = 



-4.17 + 0-0717 (age) 



20 



40 



Cohort B 

 Log e weight 



z 2 = 0.88 



n = 47 



60 



80 



100 



: -4.74 + 0.0765 (age) 



20 



40 



60 



80 



100 



Cohort C 



Log e weight = -4.55 + 0.0748 (age) 



20 



40 



60 



80 



100 



Cohort D 



Log B weight = -4.47 + 0.0773 (age) 



^ = 0.81 

 n = 61 



20 



40 



60 



80 



100 



3 

 2 

 1 





 -1 

 •2 

 -3 

 -4 

 20 



Cohort E 



Log e weight = ^1.66 + 0.0816 (age) 



? = 0.85 

 n = 66 



40 60 80 100 



Cohort F 



Log e weight = -4.27 + 0.0697 (age) 



f 2 = 0.86 

 n = 47 



20 



40 



60 



80 



100 



Age (days) 



Figure 6 



Comparison of growth in wet-weight (grams) among six spotted seatrout \Cynoscion 

 nebulosus) cohorts collected in 1995. See text for cohort hatchdates. 



fieri versions of Equations 1 and 2 ( see above ) were fitted, in 

 which all coefficients of temperature were set equal to zero, 

 so that Equations 1 and 2 represent simple linear relation- 

 ships with age. For the first path, the estimate of slope was 

 0.153 fjm/d (P<0.0001); that for the second path was 

 -0.065 fim/d (P<0.0001>. Addition of quadratic terms to 

 each model was not supported (P=0.81 and 0.12, respec- 

 tively). For the first path, whether intercept or growth rate 

 were associated with temperature was determined by test- 

 ing whether the parameters j3 01 , /3 n , and j3 12 were equal to 

 zero. There was no evidence that any of these parameters 

 were different from zero (P=0.45, 0.35, and 0.42, respec- 

 tively); the latter two may indicate that the data do not 

 support the contention that growth rate depends on tem- 

 perature in this range (1-21 d). For the second path, tests 

 "I /'„ ,=0 and /i 12 =0 offered strong evidence that these pa- 

 rameters are different from zero (P<0.001 in each case). In 

 particular, these results suggested for the age range 22-60 

 d, otolith growth rates decrease. The extent of the decrease 

 is strongly associated with average temperature according 

 to a quadratic relationship such that growth rates were 

 more steeply decreasing with age for lower temperatures 



and then became shallower at higher temperatures. In 

 summary, for temperatures at the lower and higher end of 

 the observed temperature range, otolith growth rates for 

 the age range 22-60 d were higher than they were in the 

 middle of the observed temperature range. 



Discussion 



Growth in body length of juvenile spotted seatrout in Flor- 

 ida Bay was faster than growth of juveniles from Tampa 

 Bay (Table 6, McMichael and Peters, 1989). Florida Bay is 

 generally considered an oligotrophic system (Fourqurean 

 and Robblee, 1999). Nevertheless, seagrass beds in west- 

 ern Florida Bay, where juvenile spotted seatrout are most 

 common (Chester and Thayer, 1990), are significantly 

 more dense than beds in northwestern Florida waters, 

 slightly north of Tampa Bay (Iverson and Bittaker, 1986). 

 Increased growth of juveniles in Florida Bay could be 

 attributed to the dense seagrass beds that provide habitat 

 for epifaunal crustaceans (Holmquist et al., 1989; Mathe- 

 son et al., 1999), which are important in the diet of juve- 



