The first section of this thesis submission consists of a systematic literature review regarding the relationship between personality constructs and therapeutic alliance. A total of 17 studies met inclusion criteria and these pertained to four personality constructs; values, interpersonal style, personality organisation, and quality of object relations. Alliance was measured in a variety of ways which made it difficult to compare studies but results were divided generally in terms of patient-or therapist-ratings. Research supported the link between interpersonal style and alliance,and was limited althoughrelatively consistent regarding the correlation between alliance and the other personality constructs. There was a significant amount of variation in methodology,however, and where this was not the case it was a consequence of reuse of study data, which limits generalisability. The review identified a need for replication studies and descriptive rather than diagnostic measures, particularly regarding values and some forms of personality organisation. The second part contains an empirical research paper pertaining to the role of values in therapeutic alliance. A total of 102 patients with depression diagnoses and 19 therapists contributed data with75 matched dyads analysed. The hypothesis that there would be a mediation effect of alliance on dyad value similarity and depression outcome was not supported. There were, however,significant correlations between value similarity and alliance, and between alliance and outcome at six months. Clinical implications and future research are discussed.