SMITH ET AL.: ORGANIZATION OF NEKTON 



patterns of earlier downstream emigration were 

 observed during 1976-78 in the Cape Fear estuary, 

 N.C. (Weinstein and Walters 1981; Weinstein pers. 

 obs.), the cause of which remains unexplained. It is 

 evident from Table 2 that spot may have been more 



2000 r 



1800- 



MAR APR MAY JUN JUL AUG SEP OCT 



FIGURE 2. — Total numbers of individuals and species captured 

 in monthly tidal creek collections. Temperatures and salinities 

 are mean values recorded at each creek in each month. 



restricted in their upsteam movement in Goalders 

 Creek where salinities averaged about 2"L lower at 

 the upsteam station than at the creek mouth. The 

 ratio of upstream to downstream station catch was 

 twice as high at Blevins Creek, supporting this 

 pattern. Other species which seem to prefer a 

 specific portion of the tidal creek to shoal habitat 

 gradient included hogchoker, weakfish, and At- 

 lantic croaker which had creek-to-shoal ratios 

 (over all months) of 0.13, 0.10, and 0.02, respec- 

 tively. Moreover, these species were far more 

 abundant at upstream sites (Table 1). 



Community Composition 



A two-way coincidence table, using a similarity 

 value of 0.200 to define clusters (Clifford and 

 Stephenson 1975), was constructed in order to 

 summarize species and site relationships for 

 pooled monthly collections at each station (Table 

 3). Included in this analysis are samples collected 

 with the Wegener ring, a gear which was expected 

 to be more successful in collecting both cryptic 

 species (e.g., Gobiidae) and shore-zone taxa (e.g., 

 cyprinodonts [Cyprinodontidae] and silverside 

 [Atherinidae]). It should be pointed out, however, 

 that any comparisons between the Wegener ring 

 and trawl samples are qualitative since no at- 

 tempt was made to compare gear selectivity, effi- 

 ciency, and area encompassed by a unit effort for 

 each sampling device (Weinstein and Brooks 

 1983). 



Species group IV (Table 3) was generally the 

 most ubiquitously distributed assemblage over 

 the range of environmental factors (particularly 



FIGURE 3. — Relative densities of Leiostomus xanth- 

 urus at tidal creek (values shown are monthly means 

 for both creek stations) and shoal sampling localities. 

 May values for Blevins Creek are drawn to half-scale. 

 Values appearing above histograms are ratios of 

 creek to shoal densities. 



459 



