cooling-water flow and time. The harmonic components were the same 

 for both units and reflected cycles of 12 and 4 months (Table 7) . 



Tabln 7. St I ch lab^iiks ( h I sck ipot tad <ind thr«<splii« conblnrd); 

 rfdriou& HNPS ponltorlnq progrtmi. 



tiaa In 



of tlni«-bat<id rayroKtlon pod«1 & s<>le 



transformad cdtchss In 

 fficient* 



ror pariofi of duration m (Mon 

 autor^yrtttslon covfriclants 

 raildudi f pr«dlct«d-obi<^rv«d) 

 cooling Matvr flow rata at Indicatad Uo I t 

 nuabar of obtarv^t ion% contr i but ing to tha Xa 



The coefficients of determination (R^) of both models were high (0.88 

 and 0.86) demonstrating the reliability of the models. In addition, the 

 forecast errors were low, and few observations in 1983 were outside the 

 95% confidence limits (Fig. 9). 



In the Millstone area, sticklebacks spawn and care for their eggs 

 and young in fresh- or brackish-water nests from May through October 

 (Table 2) . Because these habitats were not sampled by any MNPS 

 programs, sticklebacks were missing from samples or found in low numbers 

 during this period. Their abundance increased in trawl and impingement 

 samples as young-of-the-year moved to deeper waters after spawning. 

 After mid-winter, their abundance increased in impingement as they 

 returned inshore to spawn. Models of impinged sticklebacks accurately 

 described the historical data, and showed that 1983 impingement levels 

 were similar to previous years. 



Menidia spp., silversides 



Two silverside species are found in the Millstone area, the 

 Atlantic silverside (Menidia menidia) and the inland silverside (M. 

 beryllina ) . Since 1980, silversides have been separated by species; 

 most were Atlantic silverside. 



Silversides were the dominant (81%) shore-zone fish taxon in the 

 Millstone area (Table 1). They ranked fourth in both impingement and 

 trawl samples but were rarely found in the ichthyoplankton programs. 



21 



