426 
Fishery Bulletin 95(3), 1 99 7 
CD CT> 05 CT> 05 05 <=> C~3 <— > O C~ > O c~ > <— > — - — 
00 CO 00 CO CO CO 05 05 05 05 05 0> CT> 05 0> 05 CD CD CD CD 
CO co of o ^ ^ in co" r~t co" of o ^ ^ iif to r~^T co of 
Sampling date 
Figure 10 
Mean number of winter flounder, Pleuronectes americanus, caught per ten 
minute tow at three sites along a depth gradient in Great Bay Estuary, 
New Hampshire. Mean depth was greatest at site 3 (mean=6.2 m) fol- 
lowed by site 4 (2.1 m) and then site 5 (1.5 m). Error bars are one standard 
error of the mean. 
petition appears to play a role in the distribution of 
ecologically similar species along environmental gra- 
dients (Connor and Bowers, 1987). In Great Bay Es- 
tuary, low salinity and intertidal flats appear to pro- 
vide at least a partial refugium for smooth flounder 
from competition with winter flounder. 
The relation between smooth and winter flounder 
changes on a seasonal basis. At times their segrega- 
tion on a spatial scale is nearly complete, whereas at 
other times, particularly April-June at site 3 and 
September-October at site 2, they overlap consider- 
ably. Competition theory predicts that niches should 
vary temporally as a function of resource abundance 
and of the population densities of potential competi- 
tors (Llewellyn and Jenkins, 1987). The predominant 
temporal pattern of niche overlap seen in studies is 
increased overlap during resource abundance 
(Schoener, 1982; Ross, 1986). The periods of great- 
