Remarks on the Geometric Quantization of the Kepler Problem