746 



Fishery Bulletin 88(4). 1990 



Figure 1 



Station locations for age-U weakfish collections 

 in Chesapeake Bay- York River, VA. Station 

 1 (0 km, bay entrance), station 2 (30 km up 

 estuary), station 3 (45 km up estuary), station 

 4 (63 km up estuary), and station 5 (75 km up 

 estuary). 



Materials and methods 



Juvenile weakfish were collected at night (Leber and 

 Greening 1986) with an otter trawl (4.9 m, 19-mm mesh 

 wings, 1.5-mm codend). In 1983, collections were made 

 weekly at four stations in the York River estuary, 

 Virginia, from 12 July to 26 October: stations 2, 3, 4, 

 and 5; 30, 45, 63, and 75 km up the estuary from the 

 Bay entrance, respectively (Fig. 1). Station 1 was added 

 at the mouth of Chesapeake Bay in 1984, while collec- 

 tions at Station 4 were discontinued (Fig, 1), In 1984, 

 collections were made every 2 weeks from 21 June to 

 16 November. Repetitive, 2-minute trawls were made 

 at each station until 30 fish were captured or until 6 

 tows were completed. Salinity and temperature were 

 recorded with a Beckman conductivity meter at the 

 bottom at each station. 



Weakfish were stored in 70% ethanol. Standard 

 lengths of preserved fish were recorded with calipers 

 to the nearest 0.1 mm. At least three scales were taken 

 from just below the midbody lateral line curve (usual- 

 ly five) and circuli were counted according to Szedl- 

 mayer et al. (In press). When three scales could not be 

 taken, the fish was not used. The highest scale circuli 

 counts from individual weakfish were used for age 

 estimations. Age was estimated, based on daily scale 

 circuli counts, plus a 26-day estimate of the time delay 

 from hatching to first circulus formation. Also, if fish 

 were <14 mm SL (prior to scale formation), age was 

 estimated by quadratic regression of size on age based 



on laboratory-reared fish (for validation of daily circuli 

 deposition and further details of this method, see 

 Szedlmayer et al. In press). Subsequently, birthdates 

 were backcalculated by subtraction of estimated age 

 from date of capture. Birthdate frequency distributions 

 were smoothed with a 3-day moving average. Cohorts 

 within a year-class were identified by local minima in 

 the birthdate frequency distributions (Graham and 

 Townsend 1985). After separation of fish into their 

 respective cohorts, growth rates were estimated by 

 linear regression of standard length on age. Migration 

 patterns within the estuary and differential habitat use 

 by cohorts were suggested by abundance modes over 

 stations and collection dates, mean ages by stations, 

 and cohort-specific age-frequency distributions over 

 stations and collection dates. 



A 0.05 level of significance was used for analysis of 

 variance and covariance. Nonparametric analysis of 

 variance by ranks of main effects (year, station, date) 

 was used to test for abundance differences (catch-per- 

 unit-effort = number/2-minute trawl tow). Analysis of 

 covariance was used to test for significant differences 

 in growth rates among cohorts. Analysis of variance 

 was used to test for age differences among stations, 

 pooled over dates. After significance was determined, 

 Student-Newman-Keuls test was used to show specific 

 differences at a 0.05 level of significance or a 0.10 level 

 if a type II error resulted at the former level (Zar 1984). 



Results 



Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NC-^A. 



Salinity variation was associated with station, while 

 temperature variation was associated with season. As 



