Multiple Travelling Salesman Problem (MTSP) is among NP problems and the generalized form of the famous Traveling Salesman Problem (TSP). Since the MTSP is able to handle more than one salesman TSP, MTSP will be more appropriate for modeling real-world situations. Most situations are related to different scheduling and routing problems. Two works must be done simultaneously in the MTSP as follows: First, allocation of several cities to each salesman and second, specifying the order of visiting cities by each salesman. That is why MTSP is more complicated than TSP. One of the methods for solving MTSP is to transform it in to standard TSP. Up to now, many algorithms have been presented for solving this problem. In this paper, through combination of gravity and Genetic Algorithm, a new Hybrid method has been presented for solving the MTSP. Experimental results showed that the proposed algorithm would lead to better solutions compared with other algorithms.