Ethogram showing results of bats attacking (A) clicking and (B) silenced Bertholdia trigona. Line widths are proportional to the percentage of encounters that include a particular transition.

Three-dimensional flight trajectories of bats (red) attacking (A,C) clicking and (B,D) silenced Bertholdia trigona (blue) exhibiting defensive maneuvers. Moths are seen flying away (A,B) and diving (C,D). Both interactions with silenced moths resulted in capture. Arrows show the direction of flight. In A and C, circles indicate the positions of bats and moths when the moth began clicking. Gridlines are each 1 m apart.

(A) Percentage of bat attacks in which Bertholdia trigona exhibited defensive behaviors, and (B) bat capture success of moths exhibiting different combinations of behaviors. Numbers over bars show the proportion of total attacks where a behavior was exhibited (A) or capture was made (B). ^***Significantly different capture success (P<0.001).

Percentage of moths (A) diving and (B) captured by bats relative to moth distance from a light source. Numbers above bars indicate the proportion of attacks resulting in a dive or capture. Frequency of moth diving and bat capture success did not differ significantly with moth distance from a light (see Results for statistics).

Box plots of (A) bat–moth distances and (B) bat echolocation pulse intervals at the initiation of Bertholdia trigona defenses and bat approaching behavior. Moth defenses were initiated at a similar range of bat–moth distances and pulse intervals, with the exception of fly-aways for silenced moths. Note that all moth defenses were initiated after bats began their approach. Box plots show 5th, 25th, 50th, 75th and 95th percentiles of distributions. ^*Significantly different distributions (P<0.05).

Directionality of Bertholdia trigona diving behavior in response to bat attack: (A) overhead before dive, (B) overhead of dive, (C) profile before dive and (D) profile of dive. Prior to diving (A,C), moths flew randomly with respect to the direction of the oncoming bat. However, the diving behavior (B,D) had a strong directional bias away from the oncoming bat. Moths often flew slightly upwards (C) prior to diving (D). Arrowheads mark the starting (A,C) and ending (B,D) positions of moth flight. Moth flight trajectories were rotated and translated such that the bat approaches from the right (black bat symbols) and the initiation of the moth dive is at the plot’s origin. Dashed black arrows indicate the median distance and direction of moth flight when the direction of flight was not randomly distributed. P-values indicate results of the Rayleigh test.