The following paper details results for the classification of Papanicolaou stained cervical smear cell nuclei, using Gray Level Co-occurrence Matrix (GLCM) and Markov Random Field (MRF) image textural features. Following imaging, cell nuclei are extracted via fast morphological gray-scale transforms. Textural features comprising seven GLCM features and eight MRF model parameters are obtained from each nucleus image. Following feature extraction, Linear Discriminant Analysis (LDA) is used to provide optimal feature sets with reduced dimensionality. Finally, a quadratic classifier is trained and used to classify a set of 117 cervical cells into normal or abnormal classes. Cross-validated classification results indicate correct classification of 88% using 5 texture features.