Previous research in this area was conducted through observations of a small pteropod mollusk, or "sea butterfly," whose locomotion in water is similar to that of a butterfly's flight. That revealed two modes of locomotion: in one, cilia mode, the organism swims forward much like a micro-organism, using waves of beating cilia, or hair-like structures; in another, flapping mode, the wings are extended and flapped back and forth in a symmetrical manner, propelling the body forward. These results showed that this particular organism was able to use both modes: one pertaining to the microorganisms, the other to the insects or birds. As the pteropods grew, observations by Childress with his colleague, Robert Dudley, a biologist at the University of California, Berkeley, showed that the wings enabled more rapid swimming. Extrapolating the data backwards to small size, it was found that wings ceased to be effective at a critical size, establishing a transition size for winged flight.
Building on this scholarship, Childress and his colleagues at the Courant Institute's Applied Mathematics Laboratory sought ways to study free flight in the laboratory. They first replicated the forward flight of the pteropod by driving a horizontal rigid blade in a vertical oscillation while immersed in fluid. The blade was mounted on a vertical shaft, free to rotate in either direction. The blade flapped horizontally according to Newton's law of motion. It was found that the transition seen in the pteropods occurred also with the flapping blade. The transition depends upon both the size of the blade and the frequency of flapping. The researchers were thus able to study the transition by varying the frequency instead of the size. Below a certain frequency the blade ceased to rotate.
To simulate the hovering flight of a flapping body, the researchers created a vertical "oscillating wind tunnel," by using a large speaker operated in the range 10-100 Hertz and driving an oscillating column of air in a vertical, cylindrical flight chamber. They then simulated a bug using a small winged body made of paper and placing it in the airflow. The wings are driven to flap and the bug hovers in the flow. This allows analysts to compare the hovering of a passive flexible body in an oscillating airflow with that of an active flapper. The researchers then measured the minimum airflow amplitude needed for geometrically similar bugs of various sizes to hover in the oscillating air and were able to show how the optimal flapping frequency changes with size.
Childress and his colleagues are presently comparing these observations of free passive flapping flight with models of insect flight. The work promises to provide a new approach to the study of flapping flight, enabling studies of free hovering of winged bodies.