The gliding ant Cephalotes atratus was the first described

example of an animal capable of sophisticated aerodynamic control in the absence

of obvious morphological adaptations for aerial behavior. In this thesis I

present a series of studies aimed at quantitatively describing the aerodynamics

of stability and control in C. atratus, using a combination of

field studies and modeling experiments.

In Chapter 1, I present a quantitative analysis of the 3-D trajectories followed

by gliding C. atratus ants, measured using multiple-camera

photogrammetric techniques in a natural rainforest environment. These

trajectories represent the first attempt to measure gliding trajectories in a

natural environment for any animal, and provide a data-driven view of variation

in aerodynamic characteristics and trajectory shape across multiple C.

atratus individuals. The 3-D analysis conducted in Chapter 1 shows

that C. atratus reach equilibrium glide speed (μ=4.12,

σ=0.59) within 1s of dropping from the canopy. Lift-to-drag ratios during

equilibrium gliding were higher for ants dropped further from a target tree

trunk, but for a given horizontal distance between drop point and target tree

trunk, lift-to-drag ratios were not observed to vary significantly with size.

In Chapter 2, I present a kinematic analysis of the use of posture by gliding C.

atratus ants. By dropping C. atratus workers into a vertical

wind tunnel and analysing their behaviors using multiple cameras and posture

estimation techniques, I show that gliding C. atratus ants

immediately adopt a parachuting pose with the legs elevated above the body axis

when dropped. Changes in gaster flexion angle as well as the fore-aft angles of

the mid- and hind-legs were observed in all analysed bouts of gliding, but

statistical attempts to find correlations between static postures and body

kinematics were ultimately unsuccessful due to the rapidity with which ants in

the wind tunnel changed posture.

In Chapter 3, I present a quantitative description of stability and control in

gliding ants, using dynamically scaled physical models to show how the postural

changes discovered in Chapter 2 result in changes in the aerodynamic forces

experienced by the ant as well as its stable body orientation. I show that the

standard gliding posture adopted by a falling C. atratus ant is

aerodynamically stable in both pitch and roll, and that subtle changes in

posture and the location of the center of mass result in significant changes in

the magnitude and direction of the aerodynamic force experienced by the ant.

The postures associated with pitching and turning maneuvers, moreover, are also

aerodynamically stable, which means that gliding C. atratus

ants do not jeopardize their stability while conducting aerial maneuvers.

Finally, by testing the aerodynamic performance of a flat, splayed-out posture

for C. atratus, as well as a modified C. atratus with

shortened legs, I show that the elevated-legs posture adopted by C.

atratus is fundamental to its stability, and that leg length is a key

predictor of aerodynamic stability and control.