This work reports experiments with four hierarchical clustering algorithms and two clustering indices for on-line handwritten characters. The main motivation of the work is to develop an automatic method for finding a set of prototypical characters which would represent well the different writing styles present in a large international database. One of the major obstacles in achieving this goal is the uneven representation of different writing styles in the database. On the basis of the results of the experiments, we claim that a good set of prototypes can be formed from the combined results of the different clustering algorithms. However, the number of clusters cannot be determined automatically but some human intervention is required.