Voyager 2's recent encounter with Uranus has brought about a renewed interest in the nature of planetary rings. It has been difficult to explain, for example, how such rings can remain narrow and discrete and not expand as a result of the dissipation of energy from collisions of particles within them. A paper in the May issue of Geophysical Research Letters offers an enthusiastic restatement of one view on this question, first proposed in 1981. Advocates of the “shepherd satellite” idea, however, are far from swayed.

The new paper, by F. Curtis Michel of Rice University (Houston, Tex.), restates his idea that particles in rings may exhibit a “coherent, rolling-type motion” such that significant collisions do not occur. Michel suggests that the ring “particles” (probably better visualized as boulders) may be so densely packed that they may be treated as a fluid. In Michel's model, each individual particle would follow an inclined, elliptical orbit that would not intersect with any other such orbit. Together, the motions of all the particles would generate a toroidal surface (Figure 1). The absence of collisions would account for the apparent paucity of small particles in the rings of Uranus, Michel says.