parameters from the 1 ength-at-age data. The w parameter (tlie product of 

 L«> and K) of Galiucci and Quinn (1979) was calculated and their 

 graphical procedure was used to compare the theoretical growth of 

 females and males. This involved plotting two standard errors on each 

 side of the point estimates of L^o and K and joining the lines to form a 

 rectangle. If the resultant rectangles had little or no overlap, it was 

 inferred that a significant difference existed between the female and 

 male growth parameters. 



Probit analysis (SAS Institute Inc. 1982b) was used to determine 

 the median (50%) length of sexually mature females. The number of 

 females reproducing in the Niantic River each year was estimated by 

 determining their abundance in each 2-cm length increment starting with 

 25 cm. Fecundity (annua] egg production per female) was estimated using 

 the length-frequency data with the length-fecundity relationship 

 described by a functional regression for Niantic river winter flounder: 



^ 4 1 ? A 7 

 fecundity = 2.4837(length) (n=49, R =0.76). 



The mean fecundity was the sum of all individual fecundities divided by 

 the number of spawning females. The sum of the fecundities gave total 

 egg production for the year. 



Estimates of survival (S) were made using the abundance and age 

 data from 1977 through 1983 with the method of Robson and Chapman 

 (Ricker 1975): 



^ N+T-1 

 where T = N^ + 2N2 + 3N^. . . 

 i: N = Nq + N^ + H^... 



Movements 



Since 1980, 4,978 winter flounder were each tagged with a Petersen 

 disc to determine their movements and exploitation by fishermen. Most 

 fish in 1983 were tagged in the Niantic River during the first week of 

 March; these fish were also branded as part of the population abundance 



