Comment on "Long Time Evolution of Phase Oscillator Systems" [Chaos 19,023117 (2009), arXiv:0902.2773]

A previous paper (arXiv:0902.2773, henceforth referred to as I) considered a general class of problems involving the evolution of large systems of globally coupled phase oscillators. It was shown there that, in an appropriate sense, the solutions to these problems are time asymptotically attracted toward a reduced manifold of system states (denoted M). This result has considerable utility in the analysis of these systems, as has been amply demonstrated in recent papers. In this note, we show that the analysis of I can be modified in a simple way that establishes significant extensions of the range of validity of our previous result. In particular, we generalize I in the following ways: (1) attraction to M is now shown for a very general class of oscillator frequency distribution functions g(\omega), and (2) a previous restriction on the allowed class of initial conditions is now substantially relaxed.