Norcross and Wyanski: Recruitment pattern and abundance of Paralichthys dentatus 



593 



August 1987, and at monthly intervals thereafter. 

 During expected periods of peak age-0 summer floun- 

 der abundance in 1987 and 1988, additional samples 

 were collected throughout the study area (Fig. 1). 



Sampling was reduced spatially and further re- 

 duced temporally in 1989. Sampling was eliminated 

 at Occohannock Creek, the site at which the fewest 

 number of summer flounder were captured. Sampling 

 was conducted April through August at the other four 

 sites. Only trawling was continued. 



We measured the total length (TL) of each sum- 

 mer flounder and used the length-frequency data to 

 identify age-0 individuals. A birthdate of 1 January 

 (Smith et al., 1981) was used when designating year 

 class, although age-0 summer flounder may have 

 been collected the preceding October through Decem- 

 ber. For each gear, data from all sampling efforts were 

 pooled by month and by year class and catch per unit 

 of effort (CPUE) was calculated as the mean num- 

 ber of age-0 summer flounder per 15-m seine haul 

 or 5 minutes of trawl sample. To make sample sizes 

 more similar among the treatment groups (year 

 class) in statistical analyses, the 15-month time 

 period over which a year class was sampled was 

 separated into two time intervals: October-June 

 and September-December. July and August data 

 were not included in analyses because of bias pro- 

 duced by the clogging of meshes. 



Some data were eliminated from statistical 

 analyses owing to changes in the gear. Only seine 

 data for the 1987 and 1988 year classes were com- 

 pared because a different seine was used in 1986. 

 Unlined trawl data for the 1986 year class were 

 eliminated because the mesh size was smaller than 

 in subsequent years. Because of nonrandom (fixed) 

 station locations and nonindependent samples, 

 nonparametric statistical tests were used to ana- 

 lyze the CPUE data. For each gear, the Mann- 

 Whitney test or Kruskal-Wallis test was used to 

 compare monthly CPUE values among years (Zar, 

 1984). If the null hypothesis in the Kruskal-Wallis 

 test was rejected, a multiple comparison test 

 ( Dunn, 1964 ) was used to determine which means 

 were significantly different. If P was <0.05, the 

 results were considered significant. 



Results 



We were able to identify the age-0 year class for 

 15 months (October through December of the next 

 year) using length frequencies from all four years 

 of data combined (Fig. 2). Table 1 shows the appli- 

 cation of these monthly size-at-age criteria to iden- 

 tify the age-0 specimens in individual years. Sizes 

 ranged from 11 mm to the largest age-0 specimen 



of 285 mm. Little to no change in mean size was ob- 

 served from October to May, whereas rapid size changes 

 were apparent from June to September. 



Though sampling effort and gear varied among 

 years, age-0 summer flounder were caught within 

 Virginia waters in each year of the study. Over the 

 four years of sampling, age-0 P. dentatus were cap- 

 tured each month but not during every month of ev- 

 ery year ( Table 2 ). Summer flounder exhibited a pro- 

 longed period of recruitment to inshore waters as age- 

 specimens were captured in Virginia estuaries from 

 October to May (Table 2, Fig. 2). Newly settled speci- 

 mens (<20 mm) were collected throughout the fall 

 and winter of 1986-87; however, they were not col- 

 lected in the fall and winter of 1987-88 and 1988- 

 89. When age-0 specimens first appeared in April of 

 1988 and 1989, they were already >20 mm. 



The highest CPUE values were reported for April 

 through September (Table 2). Comparisons of CPUE 



