Considerable data have been reported indicating that there is a relationship between suicide or more serious suicide attempts and alterations in indices of serotonin (5-HT) function in patients. The fact that these data are based on studies of both suicide completers as well as more serious suicide attempters and involve a range of experimental techniques spanning biochemical assays of postmortem brain tissue as well as neuroendocrine challenge tests and platelet studies strengthens this finding. However, whether altered serotonin indices are specifically associated with violent compared to nonviolent suicide remains less clear. The relationship may be with some other aspect of suicidal behavior such as the overall lethality of the suicide method, impulsivity, or degree of planning. Suicide method appears to be related to modeling effects or to the relative availability of different methods of suicide, which argues against a predominant role for biological factors in selecting suicide methods. The relative importance of biological vs. modeling and sociological factors in determining suicide method will require more comprehensive studies in which the contribution of all of these factors is assessed simultaneously in the same patient population.