Summary: We have developed a novel experimental model of cancer immunity in the frog, Xenopus, which may provide a useful alternative to murine tumor models and a way to assess whether the control of tumor development is a fundamental function of the immune system of vertebrates. In Xenoptis, tumor immunity can be studied in two developmentally distinct immune systems. The larval immune system reflects characteristics of an ancestral system that appears to function without classical MHC class I antigen presentation and an efficient effector mechanism. The adult system appears more highly evolved in that it is remarkably similar to that of mammals and is able to generate a potent antitumor response. This amphibian model also provides a unique system with which to investigate a postulated role of heat shock proteins as components of an ancestral system of antigen presentation and/or immune surveillance that predates the antigen presentation pathway that exclusively involves MHC molecules.