Linkages between the two studies are established by a focus on reunification and a comparison of common demographic variables. In the prison study, 50 incarcerated mothers were interviewed about their relationships with children. Mothers were also asked to participate in a problem-ranking exercise in which they rated frequency, intensity, and coping mechanism used in responding to 24 problems of separation identified from the literature. In the parole study, 187 parole records of mothers in three parole offices were reviewed. Intended and actual reunification were the dependent variables. Only 32 percent of the mothers in this study intended to reunify immediately with their children, and 28 percent indicated that it was unlikely they would ever reunify. Problems such as missing children, wanting to hold children, guilt, anger, and sadness about separation were the problems experienced most severely. In the parole study, 73.4 percent of the records evidenced reunification of mothers and children; length of substance abuse, race, prior incarcerations, and length of time incarcerated were significantly related to reunification. 38 tables and a 71-item bibliography