Due to the effect of tides on sample density, starting in 1984 sampling time was based on tidal stage. 

 During the occurrence of Stage 1 larvae sampling was conducted close to low slack and when Stage 3 and 

 4 larvae were common sampling was conducted during the latter portion of a flood tide. Prior to 1984, 

 larval abundance in the lower portion of the Niantic River may have been underestimated, as sampling 

 was not synchronized with tidal stage. 



Abundance and distribution 



The distribution of larval winter flounder near MNPS was examined based on the area-wide sampling 

 conducted during 1974 and 1975, when up to 16 stations were sampled (Fig. 3). To reduce sampling bias 

 the data were restricted to oblique samples collected with 333-(im mesh nets. The date of peak abundance 

 was estimated from the point of inflection (Equation 8) of the Gompertz function to compare the temporal 

 distribution in six areas. The grouping of stations resulted in 2 to 9 samples per week in 1974 and 2 to 

 15 samples per week in 1975 for calculating a weekly mean used to construct the cumulative density curve 

 for the Gompertz function. Assuming that the Niantic River was the primary source of winter flounder 

 larvae in the area, the progressive date of peak abundance would provide a relative measure of dispersal 

 rate from the Niantic River. All fits of the Gompertz function exceeded an R} value of 0.97. A similar 

 pattern of the time of peak abundance was found for both years (Table 21). As expected, the earliest 



Table 21. Estimated date of peak abundance for larval winter flGunder based on the 



inflection point of the Gompertz function for six areas around Millstone 

 Point in 1974 and 1975. 



Area 



Niantic River 



Mouth of the Niantic River 



Mid-Niantic Bay 



.Jordan Cove 



Twotree Channel 



onshore 



19 7 4 



19 7 5 



68 



