The fixed destination multi depot multiple traveling salesmen problem (MmTSP) in which more than one salesmen depart from several starting cities and having returned to the starting city, form tours so that each city is visited with exactly one salesman and the tour lengths stay within certain limits. This problem is of a great complexity and few investigations have been done on it before. In this paper we propose a new GA chromosome and related operators for the MmTSP and compare the theoretical properties and computational performance of the some common operators and strategies of genetic algorithms. We also compare results in problems with small dimensions to the optimal answers obtained by solving the problems by Lingo 8.