Determination of earthquake source parameters is a non-linear and multi-dimensional problem in which trade-off problems between the parameters can be significant and complicated. Accordingly global optimization is desirable to find the best fit parameters in a large model parameters space. We apply Genetic Algorithms (GA) to search for optimal source parameters in a multiparameter space. Results show that the GA method can rapidly explore a large multiparameter model space and find optimal solutions for the source parameters with a relatively small number of function evaluations. Results with the data of the 1988 Nepal-India border earthquake indicate that GA determined source parameters better explain the observed waveforms. ¿American Geophysical Union 1995