Airborne Chemical Information and Context-Dependent Post-Strike Foraging Behavior in Pacific Rattlesnakes (Crotalus oreganus)
Predators demonstrate context-dependent foraging behaviors to dynamically and successfully track prey and can use multiple cues in this process. In squamate reptiles (snakes and lizards), chemical signals from prey significantly influence predatory behavior, especially substrate and airborne cues. In this study, we examined behavioral variation in rattlesnakes (
Crotalus oreganus
) during strike-induced chemosensory searching (SICS), a sterotyped complex of behaviors seen in squamates. Rattlesnakes can use both substrate and airborne chemical cues during SICS, but we sought to determine the changes during SICS when either substrate, airborne, or air-deposited chemical cues were the only types available to snakes in a Y-maze. We hypothesize that these cues represent the spectrum of chemical information available in the natural environment. We also modified scoring of choice in the Y-maze by deriving a choice penalty score, a reflection of how extensively the snake explored the unscented arm of the maze. In the presence of substrate trails, rattlesnakes relocated prey fastest, had highest rates of tongue-flicking, and received the lowest choice penalty scores during SICS. Airborne chemical cues enabled successful relocation, but rattlesnakes took longer to relocate prey, increased the frequency of many searching behaviors, and more extensively explored the Y-maze (more negative choice penalties). When air-deposited cues were the only type available, rattlesnakes took the longest to choose an arm, had the lowest rates of tongue-flicking, and backtracked most often. We suggest that as prey odor becomes more dilute, rattlesnakes demonstrate behavioral plasticity in SICS to preserve their ability to relocate prey.
Behavioral trials were run in a modified Plexiglas Y-maze that could conduct two different scented airstreams (Parker and Kardong, 2005). Fans were inserted into each arm with laminar airflow blocks in front (striped squares) to prevent turbulent flow. The arms were marked to calculate choice penalty (distance the snake moved in the unscented arm before making a choice; negative value). Gray represents an example of the scented area of the maze, and the dashed line is the path of the snake. In this example, the rattlesnake would have received a choice penalty of –3 for going past the 3rd hatch mark in the left arm.
Rattlesnakes (n = 22) had more negative choice penalties (=worse performance; 0 to –5 scale) during post-strike trailing in the Airborne and Deposited experiments than in the Substrate experiment. In the Substrate experiment, rattlesnakes only had access to substrate trails from their envenomated mouse in the Y-maze. In the Airborne experiment, only airborne cues were available from their struck mouse. In the Deposited experiment, an airborne trail from their struck mouse was run through the maze before shutting off the airflow and allowing snakes to enter. Bars represent means (+S.E.M.; –95% C.I.). Different letters represent statistically significant differences (P < 0.05) between experiments.
Metrics and behavioral variables scored during post-strike trailing in rattlesnakes across the three experiments. Each bar represents a mean (+S.E.M.; –95% C.I.). (A) Rate of tongue-flicking (RTF) was highest in the base of the Y-maze within the Substrate experiment and across the experiments. RTF was also highest in the base of the Y-maze for each experiment. (B) From their first emergence until completion of the maze, rattlesnakes took longest to make a choice in the Deposited experiment. (C) Rattlesnakes emerged from the holding box at the base of the maze most often in the Deposited experiment. (D) While trailing, rattlesnakes turned around most often in the Deposited experiment, especially in the arms of the Y-maze. Different letters of the same case represent statistically significant differences (P < 0.05) between experiments. Asterisks represent different levels of significance (*0.01 < P < 0.05, **0.001 < P < 0.01, ***P < 0.001). Descriptions of each experiment are given in the legend for Figure 2.
Contributor Notes
Associate Editor: D. S. Siegel.