Studies concerned with the influence of personality variables on recovery from surgery are critically reviewed. There is some evidence that high levels of neuroticism or trait-anxiety are associated with more distress and slower recovery from surgery. Evidence concerning other personality variables is less conclusive, and those positive associations that have been found may be attributable to overlap between the assessments used and trait-anxiety measures. Further research should involve patient groups that are more homogeneous in terms of sex and type of surgery. Progress in the field would also be furthered by the use of factorially grounded personality measures and standard recovery variables such as pain ratings, respiratory complications or resumption of normal activities. The links which have been found between personality and clinical outcome measures can be explained in terms of effects on subjective distress, or on the patients' involvement in behaviour which may promote recovery, or on physiological and immunological mechanisms.