A maple seed. From http://commons.wikimedia.org/wiki/File:Maple-seed.jpg.
Watching a maple seed fly is an interesting experience—they flutter and twist very rapidly. Because they twist and spin as they fall, scientists say they autorotate. In fact, they autorotate quite stably—a factor that allows the wind to carry them far from their parent trees.
Maples aren't the only trees with autorotating seeds—hornbeams, for example, have similar winged seeds. In all of these seeds, the autorotation is thought to help create extra lift on the seed, enabling it to travel farther from the parent tree. (Rambling offspring are a benefit for plants, because plant seedlings compete with surrounding plants for soil nutrients, sunlight, and water. If they land too close to the parent tree, they end up competing with their own parents—which benefits neither parent nor offspring, and therefore is detrimental to the survival of the species.)
Maple seeds and other autorotating seeds produce surprisingly large amounts of lift as they fall, considering how small and relatively slow they are. This is similar to the wings of many insects, which can produce a lot of lift from a relatively small surface area. Insect wings create this lift through the production of a leading edge vortex (LEV)—that is, a maelstrom of disrupted air along the edge of the seed that is "cutting through" the air as the seed falls. (Think of a wing—one edge of it is pushing through the air as it moves forward. The other edge trails along behind. The edge cutting through the air is the leading edge.)
In the 12 June issue of Science, Lentink et al report results of an investigation into the motions of maple seeds as they fall. Because the LEVs generated by insect wings help the insects produce significant lift, the researchers reasoned that maple seeds might produce similar LEVs.
Maple seeds are relatively small, and studying them while they fall can be challenging. This is especially true if one is interested in observing the flow of air over and around the seed as it falls. Therefore, as an initial test, Lentink et al built a scale model of a maple seed that was somewhat larger than a real seed. To make studying the movement of the air over the seed easier, they attached the model seed to a large arm inside a tank of mineral oil.
It may not be immediately clear how putting a model seed in mineral oil can be used to study the flow of air around a real seed. It turns out that this works because air and mineral oil are both fluids—substances that can flow in response to stress (pressure). As it happens, all fluids behave pretty much the same way under specific kinds of stress, provided that their differences in viscosity (resistance to flow, or thickness) are taken into account. The main difference viscosity makes is in the force required to move through the fluid—as you know if you've ever tried to walk under water. The more viscous the fluid, the more force is required to push through it, and the more slowly it returns to its original position. This latter property is the reason that many fluid dynamics studies are performed in oil or water, rather than air: the higher viscosity of a liquid makes observing its flow paths much easier. The path the liquid follows around the object is the same as the path that air would follow, so the results of the study are easily transferred to air.
Lentink et al used digital particle image velocimetry (DPIV) to make an image of the fluid flow around the model seed as it "fell" through the oil. DPIV is a technique that uses laser light, high-speed cameras, and computer integration to determine the velocity (speed and direction) of the fluid moving around an object in various locations. In DPIV, tiny particles are suspended in the fluid. During the experiment, as the fluid is moving, rapid flashes of laser light shine on the fluid, making the suspended particles visible for brief instances. A high-speed camera photographs the particles during each flash. The images are fed into a computer, which analyzes the locations of the particles during each instant. Because the computer knows the location of each particle at specific instances in time, it can calculate the velocity of each particle over time. Once the computer has calculated the velocities of the particles, it can create a three-dimensional image of how they move (and, by extension, how the fluid moves).
Using DPIV, Lentink et al identified a very pronounced LEV along the model seed. To confirm that their model seed accurately represents real seeds, they placed real maple seeds in a vertical wind tunnel. They adjusted the wind speed in the tunnel so that it matched the air speed the seeds would experience as they fell. As a result, the seeds hovered in place, but still spun the same way they would if they were actually falling. They recorded the motions of the seeds as they rotated. They were also able to create images of the flow of air around the seeds. The experiments with the real seeds confirmed the results seen in the model studies: maple seeds do, indeed, produce significant LEVs as they fall.
By comparing the maple seeds to other plant seeds, Lentink et al showed that the rotation of the maple seeds, and the resulting development of the LEVs, allows maple seeds to fall more slowly than non-rotating seeds of a similar wing loading (wing loading is the ratio of seed weight to surface area). Therefore, maple trees (or hornbeam trees, or other trees with rotating seeds) can produce heavier seeds (which can contain more food for the embryonic tree), but those seeds can still travel far enough from the parent trees to avoid competition.
Maple and hornbeam trees are not the only organisms to make use of the extra lift provided by LEVs, though. Hovering insects, bats, and possibly some birds also benefit from the production of LEVs along their wing edges. It makes me wonder whether "winged" marine organisms might generate similar vortices along their wings as they "fly" through the water.
Lentink, D., Dickson, W., van Leeuwen, J., & Dickinson, M. (2009). Leading-Edge Vortices Elevate Lift of Autorotating Plant Seeds Science, 324 (5933), 1438-1440 DOI: 10.1126/science.1174196