The mechanism and kinetics of the propagation step in the radical ring-opening polymerization of methylphosphirane, -phosphetane, and -phospholane, and also phenylphosphetane, were studied via high-level ab initio calculations. It was found that radical ring-opening polymerization should occur via attack of the carbon-centered propagating radical at the phosphorus center of the ring. This is a facile process (with reaction rates of the order of 10(4)-10(6) L mol(-1) s(-1) at 298.15 K), driven by the creation of a transition structure that resembles a (relatively stable) stretched phosphoranyl radical. The reaction is fastest for the four-membered phosphetanes, reflecting the best compromise between the strain in the transition structure and the strain released by the partially broken ring bond. Though slightly slower, radical ring opening of the three-membered phosphirane should also occur, but the reaction of the five-membered phospholane is not thermodynamically favorable. On the basis of the calculations for the methyl- and phenyl-substituted heterocycles, we predict that radical ring-opening polymerization should be possible for phosphetane monomers and may also be feasible for phosphiranes, though in some situations it may be hindered by a beta-scission reaction. Since the preferred propagation mechanism involves attack by a carbon-centered propagating species, we also predict that free radical copolymerization should provide a viable route to random copolymers of phosphines with normal olefins. This in turn offers the exciting prospect of incorporating phosphine units into normal polyolefins in order to alter their polarity, metal ion binding characteristics, and fire retardancy.