The electromagnetic vortex was reported by R.A. Speciale [1]-[2]. The solution of the vector wave equation for the free space propagation of unguided, collimated Bessel beams is reexamined and new features are described. First, the set of modes is doubly indexed, nondenumerable, and overcomplete. Second, the phenomenon of mode cutoff frequency is absent. Third, axial phase velocity is superluminal, i.e. exceeds c, while energy or group velocity is subluminal, i.e. less than c. Yet unlike waveguide modes, the geometric mean of the phase and group velocities does not equal c. Einstein's constant c loses significance as "the" speed of light, and instead is an upper bound on the speed of light in free space. To the two known methods for slowing light -media and guiding structures -must be added a third, orbital angular momentum (OAM). In view of these results, certain statements in classical books on electromagnetics require emendation.