451 



Abstract.— A method for using whole 

 otohth morphometries to identify silver 

 hake ^Mer!uccills bilinearis) stocks in 

 U.S. waters of the northwest Atlantic is 

 described. Whole sagittal otolith mor- 

 phometric variables of length, width, 

 area, perimeter, circularity, and rect- 

 angularity were extracted by image 

 processing and, in combination with 

 age-specific discriminant analyses, were 

 used to differentiate two stocks of silver 

 hake: a northern stock from the Gulf 

 of Maine to northern Georges Bank 

 and a southern stock from southern 

 Georges Bank to the Middle Atlantic. 

 Further support for these groupings is 

 supported by growth rate analyses: fish 

 of the northern stock grew slower than 

 those of the southern stock, resulting 

 in typically larger otoliths for fish from 

 the northern stock. This study demon- 

 strates that whole otolith morphomet- 

 ries, specific to fish age, are useful in 

 identifying silver hake stocks and can 

 be a useful tool in identifying other fish 

 stocks. 



Distinction between silver hake 

 (Merluccius bilinearis) stocks in 

 U.S. waters of the northwest Atlantic 

 based on whole otolith morphometries 



Karen L. Bolles 



Gavin A. Begg 



Northeast Fisheries Science Center 



National Marine Fisheries Service, NOAA 



166 Water Street, Woods Hole, Massachusetts 02543 



Present address: (for K L. Bolles) Marine Research Institute 



Skulagata 4, P,0 Box 1390 



121 Reykiavik, Iceland, 

 E-mail address (for K L Bolles) karenfghafro is 



Manuscript accepted 15 February 2000. 

 Fish. Bull. 98:451-462 (2000). 



Accurate distinction between fish stocks 

 is necessary for effective fisheries man- 

 agement to prevent overfishing of in- 

 dividual stocks, to preserve the repro- 

 ductive and genetic diversity of stock 

 complexes, and to determine optimal 

 strategies for rebuilding stocks (Gush- 

 ing, 1968; Booke, 1981; Grimes et al., 

 1987; Begg et al., 1999; Stephenson, 

 1999). Integral to attaining such man- 

 agement strategies and objectives, is 

 the necessary requirement of reevalu- 

 ating stock definitions when needed, 

 particularly when the status of the 

 resource changes, or when new tech- 

 nologies become available that may 

 provide more effective stock discrimi- 

 nation tools than those already in use 

 (Begg and Waldman, 1999). Because 

 of the uncertainties surrounding the 

 stock definitions for silver hake (Mer- 

 luccius bilinearis) in U.S. waters of the 

 northwest Atlantic, such a need for 

 reevaluation of stock definitions arose. 

 Furthermore, there was no efficient 

 method for adequately discriminating 

 between silver hake stocks that could 

 also provide a fast and accurate way 

 to determine levels of stock mixing for 

 current fishery assessment and man- 

 agement. In our study, we focused on 

 these issues by incorporating the tech- 

 nology and efficiency of image anal- 

 ysis systems, in combination with mul- 

 tivariate statistical analyses, to distin- 

 guish between silver hake stocks on the 

 basis of phenotypic differences in whole 

 otoliths. 



Silver hake, (or whiting), is a princi- 

 pal groundfish species found along the 

 continental shelf and slope from New- 

 foundland to South Carolina, inhabit- 

 ing depths from shallow inshore waters 

 to those greater than 400 m (Bigelow 

 and Schroeder, 1953; Almeida, 1987; 

 Helser et al., 1995). Silver hake are 

 ecologically important both as a pred- 

 ator and prey in the northwest Atlan- 

 tic ecosystem (Edwards and Bowman, 

 1979; Bowman and Michaels. 1984), 

 and form an important part of the com- 

 mercial fisheries throughout this region 

 (Anonymous, 1998). With annual land- 

 ings averaging 16,600 metric tons, silver 

 hake are currently considered overex- 

 ploited (Anonymous, 1998). 



Two stocks of silver hake are cur- 

 rently defined in U.S. waters on the 

 basis of analysis of research survey 

 and commercial catch data, as well as 

 multivariate analysis of external mor- 

 phometric data: a northern stock (from 

 the Gulf of Maine to northern Georges 

 Bank) and a southern stock (from south- 

 ern Georges Bank to the Middle Atlan- 

 tic) (Almeida, 1987). Distributions of 

 the two stocks vary seasonally and spa- 

 tially, probably in response to temper- 

 ature and depth (Helser et al.. 1995). 

 Silver hake that compose the northern 

 stock overwinter in the deeper regions 

 of the Gulf of Maine and during their 

 peak spawning months of July and 

 August, are found on northern Georges 

 Bank and in coastal waters east of 

 Cape Cod, north to Grand Manan (Big- 



