The reliability of DSM-III-R diagnoses of personality disorders is poor and their validity has yet to be established. There is little evidence that the features of personality pathology cluster into these diagnostic entities. For these reasons, it is important to explore alternative ways of classifying personality disorders. In this preliminary study, reliable scales were developed to assess 100 personality dimensions which were systematically developed. The factorial structure underlying the dimensions was evaluated in a heterogeneous sample of 110 subjects from the general population. Sixteen components, accounting for 81.4% of the variance, were retained for rotation to oblique structure. The components were labelled social avoidance, narcissism, insecure attachment, compulsive behaviours, interpersonal disesteem, mobility, anxiousness, conduct problems, stimulus seeking, identity disturbance, self-harm, rejection, diffidence, and hypersensitivity. Two components were not interpreted because they only had one or two salient loadings.