That host resistance against experimental neoplasms is a biological reality has long been sustained by the plethora of tumor-host systems in which immunological reactions have been demonstrated. The evidence has been derived principally from the unequivocal demonstration of tumor rejection in vivo and of serological and antitumor cellular reactivity in vitro (reviewed by Herberman, 1974). Considerable progress has been made toward the identification of the principal cellular components involved which, depending on the tumor-host system, include cytolytic thymus-derived lymphocytes (CTL), antibody-dependent killer (K) cells, and natural killer (NK) cells (reviewed by Levy and Leclerc, 1977), and macrophages acting alone, or in concert with other cell types (Evans and Alexander, 1976; Keller, 1977). In the majority of these studies, effector function has been demonstrated at a systemic level, in populations variously derived from regional lymph nodes, spleen, peripheral blood, and the peritoneal cavity (Cerottin` and Brunner, 1974). The interpretation of these reactions in relation to actual in vivo events governing tumor progression or regression is difficult in the absence of comparable data on the activity of effector cells from within the tumor itself. Information on the nature of the tumor-host interaction which occurs in situ, and is probably critical to the balance of the tumor-host relationship, is just beginning to emerge.