by stage during which high rates of mortality op- 

 erate. Although Laurence (1975) demonstrated 

 that the metabolic demands of larval winter floun- 

 der increased at higher temperatures, the growth 

 rate also increased if sufficient food resources were 

 available. However, other laboratory studies 

 (Laurence 1977; Buckley 1980) have shown that 

 larval winter flounder growth rates are dependent 

 on prey availability. 



In Niantic Bay, water temperature appeared to 

 have strongly affected growth, but in the Niantic 

 River it was less clear what factor was most im- 

 portant in controlling growth. 



Mortality 



Mortality rates of larval winter flounder were 

 estimated from data collected from three Niantic 

 River stations. Data from 1983 were excluded as 

 smaller larvae were undersampled because of net 

 extrusion (NUSCO 1987). The 3.0-mm and 

 smaller size-classes were used as an abundance 

 index of newly-hatched larvae because this was 

 the approximate length at hatching and the 

 3.0-mm size-class was collected most frequently 

 (Fig. 26). The rapid decline in the frequency of 

 larvae in the 3.5- and 4.0-mm size-classes was 

 attributed to both natural mortality and tidal 

 flushing from the river. Hess et al. (1975) esti- 

 mated the loss of larvae from the entire river as 

 4% per tidal cycle and also determined that the 

 loss from the lower portion of the river was about 

 28% per tidal cycle. Therefore, the daily abun- 

 dance estimates of larvae in the 3.0-mm and 

 smaller size-classes at station C (located in the 

 lower portion of the river) were increased by a 

 factor of 1.93 to compensate for the 28% decline 

 per tidal cycle with two cycles per day. Tidal 

 studies conducted in the Niantic River suggested 

 that older larvae (Stages 3 and 4) utilized vertical 

 migration in response to tidal flow to enter the 

 river and those within the river used a similar 

 behavior to remam there (NUSCO 19S7). The 



increasing frequency of larvae in the 6.0- to 

 7.0-mm size classes was at least partially attributed 

 to this behavior, so the number of larvae in the 

 7.0-mm size class was used as an abundance index 

 of larvae nearing the end of Stage 3 of develop- 

 ment. 



For the mortality calculations, abundance in- 

 dices for newly-hatched larvae, after adjustment 

 for tidal flushing, and for larvae in the 7.0-mm 

 size class were determined by summing the mean 

 weekly abundance (three stations combined) dur- 

 ing each larval season. Survival rate from hatching 

 through Stage 3 was estimated as the ratio of the 

 abundance index of the larger larvae to that of 

 the smaller larvae. Total larval mortality through 

 Stage 3 for 1984-87 ranged from 84.6% to 96.9%, 

 which represented a mean instantaneous rate (Z) 

 of 2.58 (Table 17). These larval mortality rates 

 are only preliminary estimates until additional ag- 

 ing information is available and new simulation 

 studies are conducted to better estimate tidal flush- 

 ing rates. Previous attempts to age larvae by 

 examining otoliths with a light microscope were 

 not successful (NUSCO 1987), but an improved 

 technique developed at University of Rhode Island 

 (Dr. A. Durbin, pers. comm.) may allow the use 

 of otoliths to age winter flounder larvae. 



Post-Iarval young-of-the-year studies 

 Abundance 



Beginning in 1983, a 1-m beam trawl was used 

 weekly in the Niantic River to collect post-larval 

 young-of-the-year winter flounder from late May 

 through the end of September (NUSCO 1987). 

 Kuipers (1975) reported that a similarly-designed 

 2-m beam trawl with at least one tickler chain 

 (the 1-m trawl has two) was nearly 100% efficient 

 in catching plaice smaller than 70 mm. Kuipers 

 (1975) and Poxton et al. (1982) noted that effi- 

 ciency of a beam trawl decreased for larger young 

 in fall and winter due to changes in behavior, 



Winter Flounder Studies 



197 



