322 



Fishery Bulletin 93(2), 1995 





Potomac 1987 



Potomac 1989 



0.6 

 0.5 

 0.4 

 0.3 

 0.2 

 0.1 

 



35 0.6 



30 0.5 



25 0.4 

 20 



0.3 

 15 



10 ° 2 



5 0.1 







a ma 



~\ i i i i r 



vo a, r» wi s 



<-* »s M sa 



s 



Potomac 1988 



Upper Bay 1989 



0.4- 

 0.3- 

 0.2- 

 0.1 

 0- 



40 0.5 



0.3 



Cohort Hatch Date 



~1 I I I I i I I r 



: ^h om« » ^m k » T r~ o 



Cohort Hatch Date 



D 

 > 



Figure 5 



Cohort-specific, mean growth rates (closed bars, mmd -1 ) and stage durations (open bars, age (d)) of 

 8-mm-total-length striped bass, Morone saxatilis, larvae by hatch date in the Potomac River, 1987- 

 89, and Upper Bay, 1989. Growth rates determined by the aggregate-sample method. Error bars 

 indicate 95% confidence intervals. 



among-cohort differences of up to 4.8 mm, 3.0 mm, 

 and 4.6 mm in the Potomac in 1987, 1988, and 1989, 

 respectively, and differences of 3.4 mm in the Upper 

 Bay. Maximum among-cohort differences in predicted 

 ages at 8.0 mm were 11, 9, and 21 days in the Potomac 

 River in 1987, 1988, and 1989, respectively, and 10 

 days in the Upper Bay (Fig. 6). 



Back-calculated (r 2 =0.67; Fig. 7) and aggregate- 

 sample (r 2 =0.62) cohort growth rates were strongly 

 and positively related to the mean water tempera- 

 tures in which larvae grew. A stepwise multiple re- 

 gression analysis indicated that densities of copepod 

 nauplii, a probable prey of first-feeding striped bass 

 larvae, were negatively correlated with larval growth 

 rate, presumably because nauplii densities were 

 higher near the beginning of the spawning season 

 (Fig. 4) when growth rates were lowest. No other 

 environmental variable, including densities of other 

 zooplankton taxa or densities of striped bass eggs, 

 larvae, or Morone spp. larvae ( striped bass plus white 

 perch, M. americana), explained a significant pro- 

 portion of the variance in larval growth rates. 



Individual growth rate variability Growth histo- 

 ries of larvae within each cohort that presumably 

 had experienced the same suite of environmental 



conditions indicated that some individuals grew 

 much faster than average. Within a cohort, individual 

 larvae differed by as much as 5.0 mm in length at 20 

 days posthatch (Upper Bay 1989, cohort hatched 23 

 May), and by up to 0.33 mmd" 1 in growth rate (Up- 

 per Bay 1989, cohort hatched 23 May). The maxi- 

 mum individual growth rate of a 3-day cohort was 

 68% higher than the mean. Variance (%CV) in lar- 

 val growth rate or lengths at 20 days posthatch 

 within-cohorts did not differ significantly (Kruskal- 

 Wallis; P>0.10) among years. There was no signifi- 

 cant relationship between variance in growth rate 

 within-cohorts and the mean (P>0.20) or variance 

 (P>0.50) in daily temperatures experienced by cohorts. 

 Evidence for size-selective mortality operating on 

 larger individuals was detected in back-calculated 

 growth histories of striped bass larval cohorts exam- 

 ined from the Potomac River, 1987 and 1989, and 

 Upper Bay, 1989. To detect evidence of size-selective 

 mortality, lengths at capture of larval cohorts col- 

 lected early in the spawning season were compared 

 with back-calculated lengths at age of older larvae 

 in the same cohort collected later in the season. For 

 10 of the 11 cohorts examined from the Potomac River 

 1987 (z?=5), 1989 (n=4), and the Upper Bay 1989 

 in-2), mean back-calculated lengths at 10 and 20 



