Revisiting Einstein's incomplete theory.

Scientists have long known that Albert Einstein skipped something a century ago when he analyzed Brownian motion--the jiggling of particles in a fluid, such as pollen in water. Now, researchers using measurements of unprecedented precision have observed the discrepancy between Einstein's model and a single particle's path.

In a landmark 1905 study that helped establish the existence of molecules and atoms, Einstein chalked up Brownian motion to random collisions among particles and a surrounding mediums molecules. Starting in the 1940s, physicists began altering their theories of these motions when they recognized that, over minuscule distances and time intervals, particles move less randomly than Einstein's model predicted. But no one had tracked a lone particle's trajectory closely enough to compare it with the revised theory's predictions.

The new experiments do just that. In the Oct. 14 Physical Review Letters, Sylvia Jeney of the Ecole Polytechnique Federale de Lausanne in Switzerland and her colleagues in Germany and Texas report laser-based measurements of the motion of microscopic beads of polystyrene and silica in water. Their data reveal subtle effects of flows missing from Einstein's theory.