Partial C^0 estimate II: the proof (continued)