accurate. A few commercial returns and those from the Connecticut Department of Environmental 

 Protection (CT DEP) and research institutions were also included. The data were constrained to include 

 only fish captured after at least 90 d at large that showed positive growth; negative or zero grovrth was 

 assumed to be due to measurement error or to severe effects of tagging which retarded growth. 



Mortality and survival 



Two methods were used to estimate survival (S) and the instantaneous mortality coefficient Z ( = -In 

 S). The annual age-length keys from 1978-79 and 1981-83 were used with the length-frequency distributions 

 of all measured fish 15 cm and greater (0.5-cm groupings) to determine total number by age. Catch curves 

 were constructed and the slope of the natural logarithm of number plotted against age was used as an 

 estimate of Z (Ricker 1975). Estimates of survival were also made using the method of Robson and 

 Chapman (Robson and Chapman 1961; Ricker 1975): 



S = ^^ (5) 



where T = yVi + 2N2+ 3N3 + ... 

 Y,N=No+ Ni+ N2 +.. 



A^o ~ number of age 3 winter flounder 

 A'l = number of age 4 winter flounder 



variance = S 



Food habits 





The food items of 306 winter flounder collected in the area of Millstone Point from June 1973 through 

 November 1974 were examined. Whole stomachs were removed and preserved in 10% buffered formalin. 

 Each stomach was cut open, examined, and subjectively ranked according to fullness, and assigned point 

 values (Hynes 1950) as follows: 



100 - full; appeared unable to hold any additional ingested material 



12 



