596 



Fishery Bulletin 92(3). 1994 



February south, of Chesapeake Bay as suggested by 

 Smith (1973). 



There are several possible explanations for the 

 interannual differences in timing of recruitment to 

 the inshore waters and size at first collection: 1 ) abun- 

 dance is so low in some years that age-0 fish are not 

 encountered; 2) newly settled summer flounder are 

 utilizing habitats that were not sampled; and 3 ) sum- 

 mer flounder juveniles do not enter estuaries at the 

 same time and size in all years. 



The first explanation seems plausible for the 1988 

 year class which apparently had no October 1987- 

 March 1988 recruitment and extremely low numbers 

 in summer (Table 2). However, there also was no 

 October-March recruitment in 1988-89, yet abun- 

 dance indices in May-August 1989 were comparable 

 to those in 1987, a year with October-May recruit- 

 ment. We sampled a limited number of fixed sam- 

 pling sites, thus in years of relatively low abundance, 

 an uneven distribution of the fish would appear as 

 though recruitment did not occur. 



Newly settled summer flounder may be utilizing 

 certain habitats which were not sampled because of 

 location or gear accessibility. Habitats, such as eel- 

 grass beds, would be difficult to sample with trawl 

 and seine gear (Able et al., 1990) and therefore the 

 flounder would be unavailable to the gear. Newly 

 recruiting summer flounder are most abundant in 

 marsh creeks in New Jersery (Szedlmayer et al., 

 1992). In some years of our study, the flounder could 

 have been present in eelgrass beds or marsh creeks 

 that we did not sample. From our data, it would then 

 appear as if winter recruitment had not taken place. 



Recruitment of summer flounder juveniles may not 

 be represented by a characteristic place, time, and 

 size of fish. Able et al. (1990) suggested that some 

 juveniles utilize the continental shelf as a nursery 

 and thus enter estuaries at a larger size. This could 

 explain the apparent lack of fall/winter recruitment 

 that we observed in the 1988 and 1989 year classes. 

 Variability in time of recruitment to inshore waters 

 of Virginia observed over the three years of fall/win- 

 ter sampling in this study is analogous to that in New 

 Jersey waters. Newly recruiting summer flounder were 

 collected in southern New Jersey estuaries from No- 

 vember 1988 through May 1989, but no summer floun- 

 der juveniles were collected in the corresponding 

 months of 1987 and 1988 (Szedlmayer et al., 1992). 



Variation of year-class strength in fish populations 

 has been a topic of investigation since it was first 

 proposed by Hjort (1914), but fluctuations in year- 

 class strength must be identified before causes of the 

 variation can be investigated. Four years of sampling 

 did not provide sufficient data to define a "normal" 

 level of recruitment; however, it appears that there 



was relatively poor recruitment of age-0 summer 

 flounder to Virginia waters in 1988 compared with 

 1986, 1987, and 1989 (Tables 2 and 3), suggesting 

 that there was large interannual recruitment vari- 

 ability. This decrease in year-class strength was veri- 

 fied by catches of age-1 summer flounder in Chesa- 

 peake Bay and nearshore coastal waters one year 

 later (Desfosseet al. 1 ). The highest CPUE values for 

 the seine and trawls occurred in spring 1986, sug- 

 gesting that the 1986 year class was possibly the 

 strongest of the four year classes. Given the larger 

 mesh size in the seine and unlined trawl in spring 

 1986, these high CPUE values are probably under- 

 estimates when compared with CPUE values for the 

 seine and lined trawl in 1987-89. 



We cannot speculate what effect poor year-class 

 strength of summer flounder will have on the fish- 

 ery two to four years later because it is not known 

 what percentage of the fishable population is depen- 

 dent on Virginia nursery areas. If Virginia waters 

 are a primary nursery area, the impact to the fish- 

 ery could be great. Because our data and those of 

 Desfosse et al. 1 documented poor year classes in 

 1987-89, the Virginia Marine Resources Commission 

 (VMRC) closed the nearshore (<3 miles) trawl fish- 

 ery for summer flounder effective 1 July 1989 

 (Travelstead 2 ) as a precautionary measure to protect 

 those year classes. 



Estimation of summer flounder juvenile recruit- 

 ment is biased by small-scale distribution patterns, 

 mesh size, and gear performance under certain con- 

 ditions. Gear efficiency changes with size offish. The 

 3.2- and 6.4-mm mesh seines and trawls used in our 

 study captured smaller specimens in the winter and 

 spring than did the 15.9-mm mesh trawl, but these 

 gear become less efficient with increasing fish size, 

 probably due to increased gear avoidance. Despite 

 similar mesh sizes, the sampling effectiveness of the 

 bag seine decreased more rapidly than did the lined 

 trawl, probably because of the movement of age-0 

 summer flounder to deeper habitats at 60-80 mm 

 TL (see Wyanski, 1990). No age-0 summer flounder 

 were captured by seine later than August. We agree 

 with Williams and Deubler ( 1968) that environmen- 

 tal factors, such as current velocity and mechanical 

 clogging of nets, can also have a pronounced effect 

 on sampling success for flounder. We found that gear 

 efficiencies depend on season (e.g. density of jelly- 

 fish) and size of flounder. 



1 Desfosse, J. C. J. A. Musick, A. D. Estes. and P. Lyons. 1989. 

 Stock identification of summer flounder i Paralichthys den tat us I, 

 in the southern Mid-Atlantic Bifjht. Virginia tnst Mar Sci . 

 Gloucester Point, VA 23062. Ann. Prog. Rep. WB-86-01-03. 



2 Travelstead, J. VMRC, Newport News, Virginia 23607. Personal 

 common . .June 1<ik;i 



