O'Connell et al.: Analysis of permanent magnets as elasmobranch bycatch reduction devices 
395 
This study showed that ferrite magnets induced be- 
havioral responses in all the tested elasmobranchs and 
that permanent magnets may be able to reduce elas- 
mobranch bycatch. Similarly, O’Connell et al. (2010, 
2011) showed that permanent magnets are effective 
elasmobranch-selective repellents in field and controlled 
laboratory experiments involving tests with magnets 
and procedural controls on baited apparatuses. Robbins 
et al. (2011) concluded that magnetic deterrents in the 
form of rare-earth magnetic discs have high potential 
for reducing the bycatch of shark species that occur in 
low densities, but their use in repelling shark species 
that occur in high densities, such as the Galapagos 
shark ( Carcharhinus galapagensis), was concluded to 
be minimal. 
In addition to magnetic repellents, electropositive 
metal (EPM) repellents have also been explored for 
their ability to overstimulate the electrosensory sys- 
tem of an approaching shark (Rice, 2008; Stoner and 
Kaimmer, 2008). In both laboratory and field studies, 
EPMs were shown to repel juvenile sandbar sharks 
( Carcharhinus plumbeus; Brill et al., 2009). In labora- 
tory studies, the duration of the EPM repellency was 
short lived (-three minutes), a phenomenon attributed 
to competitive interactions among the sharks. In field 
trials, there was a 62% decrease in the capture of C. 
plumbeus with EPM hook treatments. Additionally, 
electropositive metals have been shown to deter spiny 
dogfish sharks ( Squalus aeanthias) from baits in both 
laboratory (Stoner and Kaimmer, 2008) and field experi- 
ments (Kaimmer and Stoner, 2008). Although Kaimmer 
and Stoner (2008) showed that the capture of S. acan- 
thias was reduced by 19% on hooks containing EPMs in 
the Pacific halibut (Hippoglossus stenolepis ) commercial 
fishery, Tallack and Mandelman (2009) conducted both 
laboratory and field experiments in the Northwest At- 
lantic, producing contradictory results. The reasoning 
for the contrasting findings is unclear. 
In the present study, we explore the effectiveness of 
two different permanent magnets on hooks as elasmo- 
branch repellents. We hypothesize that the capture of 
elasmobranchs would be reduced with hooks containing 
magnets in comparison with control hooks in hook- 
and-line and longline studies. Additionally, we further 
hypothesize that the presence of permanent magnets on 
hooks would not alter teleost capture because teleosts 
lack the ampullary organ. 
Methods 
Longline study 
For the present study we employed grade N52 neodym- 
ium-iron-boron cylinder magnets on 30 longline sets and 
grade C8 barium-ferrite permanent cylinder magnets 
on 54 sets in North Inlet and Winyah Bay, Georgetown 
County, South Carolina, between April and Septem- 
ber 2008. North Inlet (33°19'N, 79°10'W) is a tidally 
dominated, well-mixed estuary comprising 32 km 2 of 
mudflats, oyster reefs, tidal creeks, and salt marshes 
dominated by Spartina alterniflora (Dame et al., 1986). 
It has a mean tidal depth of 2.5 m. Winyah Bay (33°12'N, 
79°11'W) is a partially mixed estuary during periods 
of low to moderate river discharges, and a salt wedge 
estuary during higher flows. Winyah Bay averages about 
four meters in depth and has various substrate types: 
mud, sand, silt, and clay (Patchineelam et al., 1999). 
Longlines consisted of a 150-m tar-coated nylon main- 
line with 24 evenly spaced gangions (branches), each 
with a single hook. Gangions consisted of 0.75 meters 
of 317.5-kg 49-strand stainless cable and 0.75 meters 
of 226.8-kg monofilament line and were attached to the 
mainline with tuna clips. The hooks were 16/0 Mustad® 
3996 open-eye circle hooks and were baited with Atlan- 
tic mackerel (Scomber scombrus). 
The magnetic flux of the longline treatments, 
with 2.5-cm diameter, 85-g neodymium-iron-boron 
(Nd 2 Fe 14 B) and a 2.5-cm diameter, 85-g grade C8 bar- 
ium-ferrite (BaFe 12 0 19 ) permanent magnets, was mea- 
sured with a model 4048 teslameter and a transverse 
probe, model T-4048-001 (F. W. Bell, Milwaukie, Or- 
egon). The former produced a maximum flux of approxi- 
mately 14,800 gauss at the surface and were polar- 
ized through the diameter. The latter were similar in 
shape to the neodymium-iron-boron cylinder magnets 
but were polarized through the height and produced 
a maximum flux of approximately 3850 gauss at their 
surface. Before experimentation, the axis of polariza- 
tion was not assumed to be a contributing factor to 
repellent effectiveness, which is why the axes differed 
between magnets. 
An alternating experimental design consisted of mag- 
netic gangions (treatment) and control (sham-magnet) 
gangions that were characterized by having an 85-g 
lead weight similar in appearance to the magnet (Fig. 
1, A and B). Magnet-type (i.e., neodymium-iron-boron 
or barium-ferrite) was consistent for each longline set. 
Of critical importance was that treatment and control 
gangions remained separated throughout the study 
to prevent the magnetization of the control gangions. 
Magnets were attached to hooks during deployment 
and removed during retrieval. Also, to prevent the 
magnetization of control gangions for subsequent tri- 
als, the tuna clips on the magnetic treatment gangions 
were marked, allowing us to properly separate the con- 
trol and magnetic treatment gangions when not in use. 
Longlines were deployed several times each week dur- 
ing slack tides (for safety and to avoid gear tangling) in 
daylight (between 0800-1700 h) for one hour. Each long- 
line was set in a double-drape configuration with the 
use of a polyform buoy attached midway on the main- 
line. With this configuration, approximately 50% of the 
hooks (i.e., 12 hooks) rested on the substrate, while the 
remaining hooks were suspended in the water column. 
During longline retrieval, teleost and elasmobranch 
fishes were identified to species, counted, measured 
(precaudal length [PCL], fork length [FL], total length 
[TL] , stretch total length [STL]), elasmobranch sex was 
determined, and treatment type noted. 
