The present study was designed to investigate the hypothesis that female clients who tend to terminate therapy prematurely will have been assigned to a male therapist. The study also tested the hypothesis that female clients who defect from therapy will have reported a history of low self-disclosure to individuals of the same sex as their therapist. Neither hypothesis was supported by the results of this study, but findings suggest a possible bias in the manner by which male and female therapists select their clients for therapy. It also appears that female defectors may be over-identifying with their family of origin or that they may be overly dependent on it as a resource system. This may be the reason for their apparent difficulty in developing a prototype that will accommodate their therapist.