Figure 1.
The unusual anatomy of the tentacled snake's head under the scanning electron microscope.
The scaled tentacles are sensitive mechanoreceptors that respond to water movements [5]. These appendages likely aid snakes in locating fish at night, with the eyes the playing the dominant role under lighted conditions.
Figure 2.
Tentacled snake hunting posture and strategy.
A. When fish approach the concave area between head and body, the snake feints with its body before striking, generating a water disturbance that usually startles fish toward the striking jaws [12]. B. When fish approach the jaws at a right angle, the body feint usually startles fish away from the body. Adult tentacled snakes bias their strike to predict this future fish movement (C). The current study examines this behavior in naïve, juvenile tentacled snakes. Trials were included in the present investigations if long axis of the fish intersected line segment “R” in B, drawn from the tip of the tentacle through the long axis of the snake's jaw to the neck, and included strikes at both approaching and receding fish. The average intersection point for the long axis of fish on the snake's jaws is indicated at “I”, 11.6 mm from the tip of the head not counting the tentacle. The standard deviation for the intersection point was 5.6 mm for the 100 trails examined using a plastic barrier between snake and fish (see text and Figure 3).
Figure 3.
Schematic illustrating the paradigm used to prevent learning from experience.
A. Top view (camera view) of the apparatus. An upper water chamber with a bottom formed from a transparent plastic barrier contained the snake and was placed into the larger, lower chamber containing the fish. The upper chamber rests on two clear plastic supports creating a channel below the snake, separated from the snake by the clear barrier. A single fish was introduced in the channel. Thus snakes could see and strike at the fish without making contact. B. Side view showing the relationship between the two water chambers.
Figure 4.
Strikes by naïve snakes at a fish below a transparent plastic barrier.
A–C show three different strikes at fish. The upper panels illustrate the movements of snake and fish with the initial position of the fish marked in gray. Matching frames from high-speed video are shown in the lower panels. A small movement of the snake's body just prior to striking elicits a C-start escape response in the fish (despite the barrier) and the snake strikes toward the future position of the fish head. The barrier prevents contact between the snake and fish (see Figure 3 and Movie S2, clips 1–10). For ease of comparison, snake orientation may have been flipped to show a left-hand bend in the neck for these and other figures.
Figure 5.
Schematic illustration of predictive strikes by tentacled snakes.
The (mean) minimum distance between the midline of the snake's head and the midline of the fish head for 100 strikes (10 for each of 10 snakes) was 5.3 mm to the far side of the fish (illustrated as if the snake's body was to the left for all trials - see figure 2B,C). The standard deviation (3.0 mm) is indicated by the gray bar.
Figure 6.
Snake strikes at fish below a glass barrier.
A-B. Snakes did not elicit C-starts in fish below a glass barrier, but still biased their strikes to the far side of the fish (see text). The red outline is a projection of the estimated position of the fish head (see figure 7) during a C-start showing the presumed target of the snake.
Figure 7.
Strikes aimed down at the escaping fish.
A. Selected frames from high-speed video showing a downwardly aimed strike at a fish below the plastic barrier. As the fish executed a C-start escape, the snake's strike intersected the position of the moving head (see Movie S2, clips 16–25 for examples). Red outline indicates the initial position of the fish when strike began. B. Schematic illustration of the strategy and result for downwardly aimed strikes. Note that the snake's head and jaws are poorly positioned to capture a fish that has not executed a C-start, even if aimed directly at an immobile fish – i.e. the jaws could not encircle the fish along its midline. C. Relative position of fish and snake's aim for 21 downwardly aimed strikes. The gray outline indicates the initial position of the fish, which served as a reference for the X and Y axes. Point “1” marks the mean position of the tip of the fish head after 26 milliseconds measured from ten C-starts. Point “2” marks the mean position of the top, midline of the snake's head at the time it rotated down to contact the barrier, which took a mean of 26.3 milliseconds for the 21 trials. Ellipses have radii of one standard deviation for the X and Y dimensions for points 1 and 2. The red sphere, “3” has a 1 cm diameter indicating the approximate aim of the snake.
Figure 8.
Selected frames of an adult tentacled snake strike during which the head is curled around and under the body to meet an escaping fish head-on. The red outline marks the original position of the fish. These strikes were observed in the naïve tentacled snakes striking at fish below a barrier (Movie S3, clip 8) and striking at fish for the first time (Movie S3, clip 9).
Figure 9.
Handling time for swallowing fish from different directions.
Each condition included 10 trials, bars are standard deviation. Handling time was defined as the time from strike initiation until the fish completely disappeared from view from the dorsal perspective. The far right column refers to fish that were grasped tail-first, but then manipulated into a head-first position for swallowing.