A nonlinear phase tracking system driven by a baseband noise is modeled by a nonlinear random differential equation whose forcing function is a function of the tracking error in the system. The transition density of the phase is determined from the Fokker-Planck equation. Then an expansion is used in solving for the transition density using concepts of eigenfunctions and eigenvalues, and this reduces the Fokker-Planck equation to a two point boundary value problem. Under appropriate conditions, the transition density of the phase error has been obtained.