This paper presents a SLAM algorithm for a team of mobile robots exploring an indoor environment, described by adopting the M-Space representation of linear features. Each robot solves the SLAM problem independently. When the robots meet, the local maps are fused together using robot-to-robot relative range and bearing measurements. A map fusion technique, tailored to the specific feature representation adopted, is proposed. Moreover, the uncertainty affecting the resulting merged map is explicitly derived from the single-robot SLAM maps and the robot-to-robot measurement accuracy. Simulation experiments are presented showing a team composed of two robots performing SLAM in a real-world scenario.