Melvin Cohn provides a list of experiments to test predictions of his associative antigen recognition (AAR) and Tritope models in relation to standard models of immunity. The manuscript highlights important questions and some decisive experiments testing the competing models. A central aspect of good science is the ability to pinpoint the important questions, something Cohn has, and continues here, to do with great clarity. The problems posed are presented succinctly, although knowledge of Cohn's previous theoretical contributions are needed in some areas to fully understand the path he is taking here. The importance of theory, as championed by Cohn, is a message that needs repeating, as it seems increasingly that immunologists favor description over theory. I briefly comment on each of Cohn's ten experiments and then discuss in detail a critical experiment that is missing from his list-an experiment testing the timing of antigen encounter in ontogeny as a central principle of the self non-self discrimination.