Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
SAF Archives  search ask the scientists in the classroom cool science
scientists from previous shows
cool careers in science
ask the scientists

Photo of Michael Dickinson Photo of Claire Balint Michael Dickinson and Claire Balint as seen on
Life's Little Questions: How Do Bees Fly?

Click on Michael or Claire's photo to read a brief bio.

q After seeing your research on SAF I was wondering if you have ever considered conducting your experiments (the one which disorients the flies by turning them in different directions) in the dark. It would seem that dark testing would force them to use only the halteres to correct their flight instead of being combined with visual cues. Mike

A (From Michael Dickinson) We have indeed performed many experiments in the dark so that the flies can only rely on their halteres for flight stability. Such experiments tell us that haltere reflexes alone are capable of keeping a fly stable. In addition, previous researchers have examined the free flight of blind flies. Flies may be blinded genetically (there are several blind mutant strains of Drosophila melanogaster) or by covering the eyes with opaque black paint. Although their maneuverability is limited, blind flies can fly remarkably well. This contrasts with haltere-less flies which rapidly crash to the ground.

q What first made you think of doing this research to find answers to the questions about how insects fly? Royce

A (From Michael Dickinson) I began my graduate training in the field of Development Neurobiology. The project was to study how tiny sensory cells on the wing of a developing fly grow and send their axons (the long signalling potions of nerve cells) into the fly's brain and make connections with the right nerve cells. Over time, I became more interested in what these nerve cells did then how they developed. The cells turned out to function as strain sensor that tell the fly when and how much the wing bends during flight. This, in turn, got me thinking about what makes the wing bend, and I was soon traveling down the garden path of insect flight aerodynamics. As a child, I had no particularly strong interest in insects and wanted to be an artist (Rodin and Singer-Sargent were my heroes). It goes to show that you need to keep an open mind when you go off to college.

q What have you learned so far about how the fly's gyroscopes work? John C.

A The halteres work by detecting Coriolis forces. Coriolis forces result whenever something moves within a rotating frame of reference. In the case of the fly, the beating halteres are subject to a force whenever the body of the fly rotates during flight. Tiny sensors at the base of the haltere detect these forces which tend to push and pull the globular end of the haltere in and out of its narrow stroke plane. The precise manner in which the forces act depends upon the direction of rotation. In this way, the haltere sensors can differentiate among yaw, pitch, and roll. The haltere sensors are connected to wing steering muscles that change the way the wing moves in order to counter-act the imposed rotation.

q I apologize for asking a somewhat mundane question, but I am interested in the source of your UV sensitive adhesive. John M.

A (From Michael Dickinson) The UV activated glue is called "Crystal Clear" and it is manufactured by Loctite corporation. Unfortunately, it is very difficult to find in stores. Much of my supply comes from Germany, where it is sold as "Glaskleber."

q Can you see the gyroscopes on the flies with the bare eye? Gary

A (from Michael Dickinson) The halteres are indeed visible with the naked eye. Have a look at the standard collections of dried insects that accumulate around the windowsill and you'll probably see a blowfly, gnat, mosquito, or some Dipteran with a nice set of halteres. Look posterior and ventral to the forewings, just anterior to where the abdomen starts. The halteres look like dumbbells or drumsticks - a round ball sitting at the end of a narrow stalk. In some flies, the halteres are protected in a cone-like outpocket of the forewing called a calyptere. In other flies, the halteres are  naked  and therefore more prominent. The halteres are especially large in craneflies, harmless but intimidating flies that look like giant mosquitoes. Craneflies are often seen at dusk in the spring when the ground is still moist. The halteres are so large in craneflies that, if you look carefully, you can actually see them beating in a flying animals. The best view of a halteres, however, is through a stereomicroscope or good quality hand lens. At higher magnification, the affinities between the wings and halteres are quite clear.

q Aren't the 3-D vortexes that created lift for the bee also responsible for lift production in conventional, flat-bottomed airplane wings? Can you suggest a teaching reference on aerodynamic lift? T. Moran

A (from Michael Dickinson) Vortices are essential for all forms of flight, from 747s to butterflies. However, one important difference between insects and airplanes is whether the vortex is  bound  or  detached . On a standard airplane, the vortex is centered on the wing itself, so that there is a net circular flow of air around the wing called  circulation . On insect wings, the center of the vortex (termed the  core ) resides above the surface of the wing. This configuration is less stable, but it generates much more lift. There are a few exotic aircraft that make use of detached vortices, the most famous of which is the Concorde. Like insects, the Concorde makes use of extensive axial flow (air flowing outward toward the wing tip through the center of the vortex) to stabilize the vortex during flight. Without this stabilization, the vortex would grow too large, become unstable, and the aircraft would stall.

An excellent introductory book on the topic of insect aerodynamics (and Biological Fluid Mechanics in general) is  Life in Moving Fluids,  by Steve Vogel, Princeton University Press. I have written a shorter and more condensed article, entitled  Unsteady Mechanisms of Force Generation in Aquatic and Aerial Locomotion,  published in American Zoologist, 36, pages 537-554.

q Does the public have access to your high speed footage of insects in hover? If so, where could we see it? (asked by several viewers)

A (from Claire Balint): The footage is available on request, but not generally available directly from the library or the web. For something a little different, you can visit Dr. Jim Marden's website at to see a QuickTime movie of surface-skimming stoneflies. There are also several PBS specials like "Alien Empire" which show footage of insects flying.

q How do the hummingbird's wings flap so fast? How many times do the wings flap per second? Bob

A A: (from Claire Balint): Hummingbird wingbeat frequencies go up to about 80 per second. Hummingbirds can flap their wings quickly mainly because of their small size. There is a general trend for shorter appendages to be able to move more quickly than longer appendages. As an illustration, you will notice that you can wiggle your finger faster than you can beat your arms. Therefore the question becomes, do hummingbirds beat their wings at a higher rate than you would expect from their size. The answer appears to be no. The flapping frequency appears to be a function of their wing length.

However, hummingbird muscles do any extraordinary amount of work to be able to beat the wings quickly, over long arcs and for long durations. This requires very high metabolic rates: hummingbirds have the highest among warm-blooded vertebrates. The muscles are able consume lots of oxygen quickly because they have many mitochondria which are clustered around the many capillaries in the muscle. In addition, hummingbird hearts, lungs, blood and digestive systems all have special characteristics which contribute to their ability to sustain extremely high metabolic rates.

A good review of the specializations of hummingbird physiology is in this paper: Suarez, R.K. (1992). Hummingbird flight: Sustaining the highest mass-specific metabolic rates among vertebrates. Experientia 48: 565-570.

Dr. Greenewalt has assembled a lot of information about hummingbird wingbeat frequencies and some beautiful pictures: Greenewalt, C.H. (1960). Hummingbirds. Doubleday, NY.

q I wasn't quite clear about some of the information presented in the How Do Bees Fly story. Do bees also have halteres that help them fly? Do all insects have halteres? Kelly

A (From Michael Dickinson) Only flies have true halteres. In fact, the scientific term for flies, diptera, means "two wings." Most insects, including bees, have two pairs of wings for a total of four. In flies, the hindwing pairs have been transformed through evolution into the halteres. In many groups of insects one set of wings may be greatly reduced in size (such as the small hind wings of bees), but only flies have transformed one set into a sophisticated gyroscope.

The only possible exception to this rule is the Stresipterids, strange ectoparasitic insects that have transformed their FOREWINGS into a set of haltere-like gyroscopes. Some researchers think this transformation is evolutionarily independent of flies, others think that Stresipterids are flies that have "swapped" wings through some developmental mutation.

q Does the glue you use on the flies harm them? What do you do with the flies when you are done? Jeff

A (From Michael Dickinson) The glue that we use to tether flies was found by a colleague in Germany, Reinhard Wolf. Both he and I have done many control experiments to make sure that the glue does not have toxic effects on the flies. The glue is sold for domestic use as "Crystal Clear" for use in repairing broken glassware. Presumably, it has been tested carefully for its potential effects on humans as well! The ultraviolet light that we use to polymerize the glue is a bit more problematic, since it temporarily blinds the flies, much as walking in bright sunlight blinds humans until our photoreceptors adapt. For this reason, we let flies recover for at least an hour before testing.

Once glued, it is very easy to remove the tether and free the flies, and we can do so at the end of each experiment. In fact, this often happens spontaneously before the experiment is completed! In many cases, however, we need to make careful morphological measurements of the wings and body. In these cases, we sacrifice the animals by anesthetizing them in carbon dioxide, and then preserving them in alcohol.

q In creating the "virtual world " for the fly experiments, you have imagined looking through their eyes. Is their world always seen black and white? If so, what makes their eyes see so differently? Susie

A (From Michael Dickinson) Like many insects, flies are most sensitive to green light. This means that they would see their world as "black and white," in that they can't see the multiple colors required to reconstruct a color image of the world. They do, however, have specialized cells that enable them to see ultraviolet wavelengths.

It is difficult, but intriguing, to imagine seeing the world as a fly might. First, flies don't have nearly the same visual resolution that we you have to imagine a fuzzier image. Second, fly eyes are faster than our own and are very sensitive to motion. Fluorescent lights that appear constant to us, look like flashing strobe lights to flies. Objects that are stationary would rapidly "fade" from view, while moving edges would appear very bright. This different design....emphasizing temporal resolution over spatial resolution makes sense for an animals that is moving very rapidly through its world.


Scientific American Frontiers
Fall 1990 to Spring 2000
Sponsored by GTE Corporation,
now a part of Verizon Communications Inc.