Although the principles of cyclopolymerization have been well established, certain aspects of the originally proposed mechanism still remain to be justified in terms of experimental observations. It is the purpose of this paper to discuss these conflicting aspects, to present new information, and to point out those areas where no single mechanistic explanation is presently sufficient. The areas to be considered are: (1) much evidence is available that the cyclization step proceeds via a fast concerted reaction, while kinetic studies on a limited number of monomers indicate that the activation energy for cyclization is greater than that for linear propagation; and (2) although six‐membered rings are predicted in radical cyclopolymerization of 1, 6‐dienes, based upon radical stabilities, numerous examples of such polymerizations are now known to lead to large fractions of five‐membered rings; in addition, extensive stuides on model compounds leading to small molecules have shown that the five‐membered ring is predominant. New evidence indicates that a driving force for cyclization may be that the cyclized polymer is in a lower free energy level than the monomer; also, that radical stability exerts a marked influence on the ring size. This paper also includes the results of a study designed to determine the effect of the nature of a substituent at the 5‐position of 5, 6‐unsaturated radicals on the ring size of the cyclic product. Thus, it was shown that 2‐methallyloxy ethyl radical yielded five‐ to six‐membered rings in the ratio of 30:1, while the 2‐(2‐phenylallyloxy)ethyl radical yielded products in the ratio of 1:2. The latter structure which simulates cyclopolymerization of di(2‐phenylallyl)ether reflects the stabilizing influence of the phenyl group on the intermediate radical. The rate constants and energies of activation for the competing reactions were determined.