Over the last 25 years, the specialty of training and supervision has developed within the field of family therapy. Standards, policies, and learning objectives have been designed for the accreditation of family therapy training programs and for qualifying family therapy supervisors American Association for Marriage and Family Therapy AAMFT , 1991, 1993). Descriptions of training programs in various contexts have been published. Considerable attention has been devoted to designing supervision and training modalities such as live supervision and videotape review. Numerous training models and supervision approaches have evolved. Research on the outcome and effectiveness of family therapy training and supervision has slowly begun to accumulate. Finally, a number of literature reviews have been published that chronicle the above events. Although we summarize previous literature in this article, our primary intent is to assess key developments in training and supervision, outline how far we have come, and propose what we view as logical steps for continued growth in this specialty area. In this review, a distinction is made between training and supervision. Training encompasses all factors related to disseminating knowledge, including clinical supervision, curriculum development, and classroom instruction. In contrast, supervision is a more specific means of transmitting knowledge, skills, and attitudes through a relational process that entails direct oversight of trainees' clinical work (Bernard & Goodyear, 1992). HISTORICAL OVERVIEW Family therapy training and supervision has progressed through several eras (Liddle, 1988a, 1991). The first, comprised primarily of the 1970s, was characterized by a burgeoning, albeit scattered literature that addressed several issues. These included: (a) descriptions of techniques and modalities (e.g., live supervision, videotape review; cf. Bodin, 1972; Cohen, Gross, & Turner, 197G; Montalvo, 1973; Napier & Whitaker, 1972; Sonne & Lincoln, 1964; Whitaker, 1971), (b) trainees' personal therapy and the study of their own family of origin (Guerin & Fogarty, 1972; Guldner, 1978; Nichols, 1968), (c) descriptions of training programs (cf. Berman & Dixon-Murphy, 1979; Constantine, 1976; Everett, 1973; Flomenhaft & Carter, 1977; Garfield, 1979; LaPerriere, 1979; Mendelsohn & Ferber, 1972; Nichols, 1979), and (d) the influence of different variables within the training context (clinical settings, trainers' professional disciplines; cf. Erlich, 1973; Framo, 1976; Haley, 1975; Liddle, 1978; Malone, 1974; Martin, 1979; Meyerstein, 1977; Shapiro, 1979; Stanton, 1975). During this time, a number of seminal contributions provided direction for the field. One such development was the definition of three essential family therapy skills outlined by Cleghorn and Levin (1973): (a) perceptual skills, the ability to see and describe accurately the behavioral data of the therapy session; (b) conceptual skills, the ability to translate clinical observations into meaningful language; and (c) intervention skills, in-session behaviors that allow trainees to modify family interactional patterns. Another development was the specification of training models based upon major schools of family therapy (cf. Barton & Alexander, 1377; Beal, 1976; Haley, 1976). This development signaled the beginning of an axiom that continues today, namely, that the methods, theories, values, and skills taught to trainees parallel developments in the broader family therapy field. A third focus to emerge was the illumination of the modalities (live, videotape, group) that would become a hallmark of family therapy training (cf. Beavers, 1985; Berger & Dammann, 1982; Birchler, 1975; Bodin, 1972; Montavlo, 1973; Olson & Pegg, 1979; Stier & Goldenberg, 1975). As Beavers (1985) would later note, these modalities opened the training system by providing direct observation and more immediate interaction between supervisor, therapist, trainee, and client. …