Casazza and Ross: Fishes associated with pelagic Sargassum and open water off North Carolina 
349 
Figure t 
Surface sampling sites for fishes collected during summer and fall of 
1999-2003 off North Carolina. Collections with neuston nets (triangles) 
and supplemental gears (circles) (i.e. , nightlighting, dip net, hook-and 
line, longline) from Sargassum (closed symbols) and open-water (open 
symbols) habitats. The white circles with a black center off Cape Hat- 
teras represent sites where underwater video recordings were taken. 
The relationships between the quantity 
of Sargassum and species richness and 
abundance or biomass of fishes are high- 
ly variable. Dooley (1972) and Fedoryako 
(1980) found no correlation between num- 
bers of fishes and quantity of Sargassum, 
but significant positive correlations be- 
tween fish abundances and quantity of 
algae have been found in other studies 
(Moser et al., 1998; Wells and Rooker, 
2004). Sampling methods may substan- 
tially influence these results. Neverthe- 
less, it is clear that objects floating in 
the ocean, like Sargassum, attract and 
concentrate fauna (Kingsford, 1992). 
Despite several surveys and the wide- 
spread occurrence of Sargassum habitat, 
the fishes associated with this habitat 
have not been extensively documented 
along the U.S. east coast. Our objective 
was to describe fish community structure 
(species composition, day versus night 
species composition, sizes, and Sargas- 
sum abundance and fish distribution 
relationships) within Sargassum and 
open-water habitats off North Carolina. 
In addition, behaviors of fishes with- 
in and below Sargassum habitat were 
documented to better describe the close 
associations of fishes with the habitat, 
the different types of habitat usage, and 
to provide a three dimensional view of 
the distribution of fishes within and beneath the Sar- 
gassum. Our approach was to use consistent methods 
(supplemented by further sampling) and extensive tem- 
poral (diel) and spatial sampling across ocean surface 
habitats with no Sargassum to those with high densities 
of Sargassum to examine the relative contribution of 
Sargassum to oceanic fish communities off the south- 
eastern United States. 
Materials and methods 
Sampling 
Surface waters were sampled during 2-7 August 1999, 
20-27 July 2000, 22-28 August 2001, 20-26 September 
2001, 6-15 August 2002, and 17-26 August 2003, in the 
Gulf Stream off Cape Hatteras, Cape Lookout, and Cape 
Fear, North Carolina (Fig. 1). The primary sampling 
device, a l.lx2.4-m neuston net (6.4-mm mesh body, 
3.2 -mm tailbag), was towed in the upper meter of the 
water column at slow speeds (<3.7 km/h) for 30 minutes 
in 1999 and for 15 minutes during 2000-03. Sampling 
was conducted throughout the 24-h period to compare 
daylight (0700-2000 h eastern daylight time [EDT]) 
and nighttime (-0700 h EDT) collections. The neuston 
net was towed in both open water without Sargassum 
and in waters with varying amounts of Sargassum. 
Sargassum is in constant motion in the Gulf Stream, 
and without aerial surveillance its distribution and den- 
sity are unpredictable, especially at night. Because we 
were unable to consistently target a particular habitat, 
and thus balance sampling effort, the nets were pulled 
through unknown habitat and the sample was classified 
later depending on whether Sargassum was present in 
the sample or not (see Data analysis). In most cases it 
was also not possible to determine the proximity of one 
habitat to another. When Sargassum was abundant, 
the neuston net was towed directly through the clumps, 
mats, or weedlines, but on some occasions, Sargassum 
was collected opportunistically. Fishes were sorted from 
the algae in the neuston tow catches, and the Sargassum 
was weighed (wet weight) to the nearest 0.1 kg and dis- 
carded. Because neuston net tows in 1999 were of longer 
duration and catches were not handled consistently, the 
catch data from the 1999 neuston net tows were not 
analyzed statistically (see Data analysis). 
Additional collection methods supplemented the use of 
neuston nets, especially when Sargassum was too dense 
for use of the neuston net. When conditions were favor- 
able (low wind and waves), stations with nightlighting 
were sampled by allowing the vessel to drift with the 
current or maintain its slowest speed into the current. 
The deck lights of the vessel, plus two 500-W and one 
1000-W spotlights, were used to illuminate surface 
waters around the stern and both sides of the vessel, 
