When I was a grad student, the math departments of Harvard and MIT occasionally had a joint lecture for two hours in the evening. The speakers were sometimes local faculty and sometimes visitors. The speakers were allowed to talk about any topic they wanted, but they were not allowed to even mention any of their own theorems. With no chance to "show off," they produced some great talks. I particularly remember a lecture by Hartley Rogers on recursion theory, starting from zero and ending, two hours later, with a perfectly understandable presentation of a finite-injury priority argument.