Lowerre-Barbieri et al.: Age and growth of Cynoscion regalis 



645 



To determine whether the population growth rate 

 was representative of the true growth rate (i.e. 

 whether there was not size-selective mortality within 

 year classes), size at first annulus formation was 

 evaluated for sectioned otoliths from fish ages 1-12 

 (Ricker, 1975). Otolith radius to the first annulus 

 (distance from the nucleus to the proximal edge of 

 the first annulus) was measured by using a Via- 100 

 camera and monitor system with a dissecting micro- 

 scope at 24x (Lowerre-Barbieri et al., 1994). Mea- 

 surements were taken on 403 Chesapeake Bay fish 

 collected in 1989 and 1992-93 and on 47 Delaware 

 Bay fish from 1992 to 1993. Given the strong rela- 

 tionship between otolith radius and fish total length 

 (Lowerre-Barbieri et al., 1994), size of the otolith at 

 the first annulus was considered an indicator offish 

 size at age 1. A one-way analysis of variance (ANOVA) 

 was used to determine whether otolith size at first 

 annulus was significantly different by age. 



Growth was evaluated by using nonlinear regression 

 (Marquardt method) to fit the von Bertalanffy model 

 (Ricker, 1975) to observed, individual lengths of Chesa- 

 peake Bay fish ages 1-12. To remove seasonal effects, 

 only fish collected in April and May were used for cal- 

 culations. These months are the period when 1 ) somatic 

 growth rate increases; 2) otolith annuli form; and 3) 

 the largest range of sizes and ages occur in Chesapeake 

 Bay. Finally, to examine differences in growth by sex, 

 observed mean size at age in Chesapeake Bay was cal- 

 culated for each sex and compared by using a Mest. 



Figure 1 



Transverse otolith section of an age-17 weakfish, Cynoscion regalis, caught in 

 May 1985 in Delaware Bay. Arrows indicate annuli. 



Linear regression was used to determine a SW-TL 

 relationship on log-transformed data from fish col- 

 lected in Chesapeake Bay. To include the greatest 

 possible range of sizes ( 188-875 mm TL and 71-6,137 

 g SW), data were pooled over gears ( pound nets, haul 

 seines, and gill nets). A i-test was used to determine 

 if the slope of the SW-TL regression was signifi- 

 cantly different from 3 — a slope of 3 indicating iso- 

 metric growth. When only TL was given in the his- 

 toric literature, conversions were made by using a 

 TGW-TL relationship based on fish collected in April 

 and May in Chesapeake Bay 1989-93, ranging from 

 20 to 6,276 g TGW and from 140 to 875 mm TL. This 

 same relationship was used to estimate TL for cita- 

 tion-size fish. 



All data were analyzed by using statistical meth- 

 ods available in SAS ( 1988). Model assumptions were 

 evaluated by examination of residuals (Draper and 

 Smith, 1981). Rejection of the null hypothesis was 

 based on an a level of 0.05, unless otherwise noted, 

 and F-tests in ANCOVA were based on type III sums 

 of squares (Freund and Littell, 1986). 



Results 



Size and age composition 



Most weakfish collected from Chesapeake Bay com- 

 mercial fisheries during 1989-92, excluding those 

 targeted for their large size, were 

 200-600 mm TL (98%) and ages 

 1—4 years (97%). However, ob- 

 served sizes ranged from approxi- 

 mately 200 mm TL to 850 mm TL, 

 and observed ages ranged from 1 

 to 8 years (Fig. 2). The smallest 

 fish (=200 mm TL) collected from 

 market grades were similar each 

 year. However, the largest ob- 

 served fish varied from approxi- 

 mately 650 mm TL in 1990 to 850 

 mm TL in 1989 and 1992. In 1990, 

 a larger percentage (78%) of small 

 (<300 mm TL), young weakfish 

 were collected and no fish were 

 older than age 5 (Fig. 2). Most of 

 these small fish (<300 mm TL) 

 were collected by haul seine and 

 pound net (Fig. 3), whereas gill 

 nets caught fish primarily in the 

 300-400 mm TL range. 



Weakfish were not fully re- 

 cruited to market grades until age 

 2. Young-of-the-year and yearling 



