OBJECTIVE The objectives of this report are to review the clinical practice of Structural Integration (SI), an alternative method of soft-tissue manipulation and sensorimotor education, and to summarize the evidence to date for mechanism and clinical efficacy. METHODS The author's personal knowledge of SI literature, theory, and practice was supplemented by a database search, consultation with other senior SI practitioners, and examination of published bibliographies and websites that archive SI literature. RESULTS SI purports to improve biomechanical functioning as a whole by progressively approximating specific ideals of posture and movement, rather than to treat particular symptoms. Hypothesized mechanisms at the level of local tissue change include increases in soft-tissue pliability, release of adhesions between adjacent soft-tissue structures, and increased interstitial fluid flow with consequently improved clearance of nociceptive potentiators. Hypothesized mechanisms for more global changes include improved biomechanical organization leading to reductions in mechanical stress and nociceptive irritation, a perception of improved biomechanical efficiency and coordination that generalizes to the self, and improvements in sensory processing and vagal tone. Emotional catharsis is also thought to contribute to psychologic changes. Limited preliminary evidence exists for improvements in neuromotor coordination, sensory processing, self-concept and vagal tone, and for reductions in state anxiety. Preliminary, small sample clinical studies with cerebral palsy, chronic musculoskeletal pain, impaired balance, and chronic fatigue syndrome have reported improvements in gait, pain and range-of-motion, impaired balance, functional status, and well-being. Adverse events are thought to be mild and transient, although survey data are not available. Contraindications are thought to be the same as for massage. CONCLUSIONS Evidence for clinical effectiveness and hypothesized mechanisms is severely limited by small sample sizes and absence of control arms. In view of the rapidly increasing availability of SI and its use for treatment of musculoskeletal pain and dysfunction, more adequate research in warranted.