INTRODUCTION

Tiger sharks (Galeocerdo cuvier) are apex predators that feed on a variety of fish (Heithaus, 2001).  Due to this, once a prey stimulus is encountered, G. cuvier relies heavily on olfaction and electroreception.  Several centimeters prior to attacking its prey, G. cuvier exposes its nictitating membrane, to protect its eye from damage while feeding (Frazzetta & Prange, 1987).  During this time, it is thought that only its electric sense is being used to capture and kill its prey item.  The ability of this shark to detect the electric impulses from its prey items occurs through the sharks electroreceptors known as the Ampullae of Lorenzini.  The sensory cells within this electroreceptors detect a potential difference between the ampullae and the prey item.  Once this potential difference is recognized an impulse is sent to the brain via the afferent neurons.  This will allow the shark to home in on the prey item and feed (Kalmijn, 1971; Murray, 1962).

We placed one juvenile (129 cm), female Galeocerdo cuvier  in a pen that was divided by our 3-dimensional net-like apparatus.  At one side of the apparatus, 30 cm away from the edge of the pen, a 70 cm x 70 cm hole was cut with no stimulus surround the hole (control).  On the other side of the apparatus, 30 cm away from the edge of the pen, another 70 cm x 70 cm hole was cut with magnets surround the hole (treatment).  With this experimental setup, one can assume that these sharks will enter the control hole a greater amount of times than the treatment hole.  Also, one may also predict that the sharks will display a greater amount of avoidance behaviors in response to the treatment hole.

MATERIALS AND METHODS

To capture our shark, a longline was set on the Southeast point of South Bimini, Bahamas.  This longline stretched 200 m and contained 15 baited gangions.  Pieces of barracuda were used as bait.  Every 4 hours this longline was increase the survivability of any captured sharks.  Once the G. cuvier was captured, a team consisting of 4 members secured the shark for data collection. The shark was pulled alongisde the boat using a gaff to secure the line.  Once the shark was brought to the boat, all necessary measurements (PCL, FL, TL) were taken in the water to minimize the stress on the shark.  Once these measurements were taken, the hook was removed and a large dip net was placed under the shark.  During this time, a 1.5 m diameter holding tank was filled with fresh seawater aboard a 16 ft Carolina Skiff.  Once the shark was captured in the dip net, two members lifted the shark into the holding tank and once in the tank, the net was carefully removed.  The shark was then immediately taken to the holding pen and gently released into the pen. 

Due to the difficulties of successfully containing a G. cuvier in captivity, this shark could only be kept for a duration of 24 hours.  The following morning, twelve hours after the shark was captured, the shark was moved to the experimental pen with the use of two dip nets.  Once the shark was placed into the pen, the shark was observed for a period of 6 hours.  This shark was placed in a 7.5 m experimental diameter pen that was divided by a fence-like apparatus that contained two square openings (Figure 4).  The square openings were located 15 cm from the ocean floor and 30 cm away from the edges of the pen.  Also, the openings were square (70 cm sides) and were surrounded by PVC pipe to make the holes more visible to our shark.  The treatment square consisted of sixteen ceramic magnets (15.24 cm x 10.16 cm x 0.64 cm) uniformly spaced around the perimeter in groups of two (i.e., each group consisted of two magnets) . 

 To begin the trial, we began to record data as soon as the shark stopped accelerating.  We noted the number of times the shark entered the square holes along with a variety of other behaviors.  An entrance was recorded when the shark passed completely through the square opening.  Other behaviors that were noted were:  Approaches, 90¥ turns, 180¥ turns, accelerations, and bumps.  Approaches were recorded when the shark swam directly at the square or when the shark swam along the contour of the pen and then began to swim into the hole (i.e., sharks head broke the plane of the square).  There were two types of turns that were recorded, 90¥ turns and 180¥ turns.  A 90¥ turn was recorded when the shark swam towards the square and made a right-angled turn at a distance equal to or less than a half meter from the square.  A 180¥ turn was recorded when the shark swam towards the square and made a complete u-turn at a distance equal to or less than a half meter from the square.  Accelerations were recorded when the shark was within a half meter of the control or treatment square and quickly increased its speed in a direction away from either of the squares.  Bumps were recorded when the shark made contact with the edges of the square using its head. 

PRELIMINARY RESULTS (OCTOBER 2006)

Two, 3 hour trials were conducted and all behaviors of the shark were noted.  Approaches and bumps were more frequent towards the control square as compared to the treatment square.  Accelerations away from, complete avoidance, 90¥ turns and 180¥ turns were more commonly observed towards the treatment square.  Lastly, the shark entered both the control and the treatment squares an equal number of times.

Figure 1 - Total behaviors displayed by G. cuvier during the 6-hr observation period.  Avoidance behaviors consist of 90¥ and 180¥ turns, and accelerations away from control or treatment squares.

Figure 2 - The frequency of approaches (Figures 2a and 2c) to the frequency of entrances (Figures 2b and 2d) of each trial.  Fig. 2a and 2b relate to the first 3-hr trial.  Fig. 2c and 2d relate to the second 3-hr trial.

Figure 3.  The entrance, 90° turns, and 180° turns towards the control and the treatment squares.  Trial one is depicted by Figures 3a and 3b and trial two is depicted by Figures 3c and 3d. 

DISCUSSION OF RESULTS (OCTOBER 2006)

While conducting the trials, it was interesting to see that as the trial progressed, the interaction of the shark with the treatment square quickly declined.  This can be seen by the data obtained from trial 2.  As seen in Figures 2c and 2d, the shark no longer approached the treatment square after it had entered through the square.  Also, after the shark entered the treatment region in trial 2, there were a total of six turning behaviors away from that treatment square, while there was only one turning behavior away from the control (Figures 3c and 3d).  These results were similar to trial 1 in that after the shark entered through the magnetic square, he then displayed five turning behaviors away from the stimulus region and only one away from the control region (Figures 3a and 3b).  So, from this preliminary data, one can assume that the shark was affected by the magnetic field and preferred to stay several meters away from the stimulus region for the remainder of the trail. 

Although it does not appear in the data, it is important to note that in trial 2, the shark entered the treatment square, stopped swimming for 2 seconds when halfway through the square, and then slightly bent her body almost as if she wanted to turn around.  This was direct evidence that the shark was affected by the magnetic field and became "disoriented."  Galeocerdo cuvier does not have the ability to make extremely tight turns and that could be one possible explanation why our test shark was not able to turn around once he entered the magnetic square. 

In conclusion, this is very strong evidence that these permanent magnets have the ability to affect tiger shark behavior.  We cannot assume that the magnets can deter Galeocerdo cuvier because only one individual was studied in this experiment.