Writer identification is carried out using handwritten text. The feature vector is derived by means of morphologically processing the horizontal profiles (projection functions) of the words. The projections are derived and processed in segments in order to increase the discrimination efficiency of the feature vector. Extensive study of the statistical properties of the feature space is provided. Both Bayesian classifiers and neural networks are employed to test the efficiency of the proposed feature. The achieved identification success using a long word exceeds 95%.