This paper presents an effective methodology to optimally reconfigure an electrical distribution network. Selective Particle Swarm Optimization (SPSO) algorithm is proposed to find the optimal combination of switches that results in a radial configuration with minimum system power loss. SPSO is a modified Binary Particle Swarm Optimization (BPSO) with selective search space. Comparative analysis of SPSO with BPSO for network reconfiguration, under four different loading conditions, namely base, light, medium and heavy, on IEEE 69 bus system is presented to demonstrate the suitability of the proposed method. It is observed that SPSO outperforms BPSO in terms of quality of solution, voltage profile, convergence characteristics and time elapsed to complete optimization process.