The minimum spanning tree clustering algorithm is capable of detecting clusters with irregular boundaries. In this paper we propose a novel minimum spanning tree based clustering algorithm. The algorithm produces k clusters with center and guaranteed intra-cluster similarity. The algorithm uses divisive approach to produce k number of clusters. The center points are considered as representative points for each cluster. These center points are connected and again minimum spanning tree is constructed. Using eccentricity of points the central cluster is identified from k number of clusters