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



329 



pensatory growth by slow-growing individuals and 

 selective mortality against larger individuals within 

 cohorts. 



If our analysis of size-selective mortality against 

 larger individuals is correct, it may be that larger 

 individuals experienced higher mortality because 

 their faster swimming speeds increased contact rates 

 with predators. Recent experimental studies (Litvak 

 and Leggett, 1992; Monteleone and Houde, 1992; 

 Pepin et al., 1992; Cowan and Houde, 1993) indicate 

 that as larvae grow and encounter different preda- 

 tor fields, their probability of being encountered and 

 eaten by some predators may increase, resulting in 

 a higher mortality rate for larger individuals than 

 for smaller individuals. Effects of selective mortal- 

 ity against smaller individuals may not become ap- 

 parent until later in the larval stage (e.g. Post and 

 Prankevicius, 1987), because it takes time for differ- 

 ences in growth rate to result in significant differ- 

 ences in size that would lower vulnerability to pre- 

 dation (Rice et al., 1993). 



We had hypothesized that mortality would select 

 against smaller individuals within cohorts, and that 

 cohorts with higher mean growth rates and highly 

 variable growth rates would contribute more poten- 

 tial recruits than cohorts with lower and less vari- 

 able growth rates (Pepin, 1989; DeAngelis et al., 

 1991; Rice et al., 1993). In our study, although mor- 

 tality rates may have been highest for the largest or 

 fastest-growing individuals, cohorts contributing 

 most to recruitment in each year were those hatched 

 near the end of the spawning season and which had 

 higher but not more variable growth rates than those 

 hatched earlier. For example, in the Potomac River, 

 1989, late-hatched cohorts grew, on average, nearly 

 twice as fast from 5-20 days posthatch as did early- 

 hatched cohorts. The high productions of late- 

 hatched, fast-growing cohorts suggest that stage 

 duration, by reducing the time that larvae experi- 

 ence high mortalities, is more important than body 

 size alone in determining potential recruitment of 

 striped bass. 



Prey densities and growth 



The failure to detect any significant influence of prey 

 density on cohort-specific growth, survival, or abun- 

 dance at 8.0 mm SL was surprising, because mean 

 zooplankton densities were highest in the Potomac 

 River and Upper Bay in years when mean growth 

 rates, GIZ ratios and recruitment indices were high- 

 est (Rutherford et al. 4 ). Laboratory, pond, and model- 

 simulation studies have demonstrated repeatedly 

 that growth and survival of striped bass larvae in- 

 crease as prey density increases (Miller, 1976; 



Eldridge et al., 1981; Rogers and Westin, 1981; Houde 

 and Lubbers, 1986; Tsai, 1991; Chesney, 1989, 1993; 

 Daniel 8 ). The strong and dominant effect of tempera- 

 ture upon larval growth rate may have obscured ef- 

 fects of prey density at the cohort-specific level. In 

 the Potomac River, 1987, and in the Upper Bay, 1989, 

 when highest growth rates were observed, zooplank- 

 ton densities were highest, increased as the season 

 progressed, and were correlated positively with tem- 

 perature. In the Potomac River, 1988 and 1989, zoo- 

 plankton densities were lower and not significantly 

 correlated with temperature, yet growth rates of 

 striped bass cohorts at similar temperatures did not 

 differ significantly from the cohort rates in 1987, in- 

 dicating that the temperature effect dominated. 



It is possible that the failure to demonstrate a rela- 

 tionship between temperature-adjusted, cohort-specific 

 growth rates and prey densities was an artifact that 

 resulted from backcalculating growth rates from mostly 

 older larval survivors. Growth rates of these larvae 

 conceivably may have been higher than growth rates 

 of larvae that died (Miller et al, 1988; Pepin, 1989) and 

 may have obscured impacts of low prey densities on 

 growth. Although for most cohorts, back-calculated 

 growth rates and lengths at age of larvae collected early 

 in the season were not significantly lower than rates 

 and lengths at age of larvae caught later, this result 

 might be artifactual, because of the potential bias on 

 back-calculated growth histories caused by nonlinear 

 changes in the otolith-body size relationship. 



We believe that recruitment level of striped bass 

 in Chesapeake Bay is essentially set by the abun- 

 dances of cohorts that survive to 8.0 mm SL. Rela- 

 tive productions at 8.0 mm SL and abundances in 

 the juvenile surveys conducted 50 to 100 days later 

 were strongly correlated in the Potomac River and 

 Upper Bay (Rutherford et al. 4 ). Other evidence that 

 striped bass recruitment is fixed during the early 

 postlarval stage (8.0-10.0 mm TL) has resulted from 

 research on striped bass in the Choptank River 

 (Uphoff, 1989) and from the Sacramento-San Joaquin 

 system (Low 3 ). Our results demonstrate that not only 

 is recruitment potential fixed by 8.0 m SL but that 

 relatively few daily cohorts contribute significantly 

 to recruitment in most years. The success of particu- 

 lar cohorts is strongly dependent upon the tempera- 

 ture regime that larvae experience between hatch 

 and 8.0 mm SL. Examination of intraseasonal dif- 

 ferences in cohort-specific growth, survival, and pro- 



Daniel, D. A. 1976. A laboratory study to define the relation- 

 ship between survival of young striped bass (Morone saxatilis) 

 and their food supply. Calif. Dep. Fish and Game, Anadromous 

 Fisheries Branch, Admin. Rep. 76-1, Sacramento, CA, 13 p. 

 Available: California Department of Fish and Game, Anadro- 

 mous Fisheries Branch, Sacramento, CA. 



