It was the spring semester of 1969 and the population geneticist Richard Lewontin had come to Harvard to deliver a talk on a new methodology that promised to revolutionize the field of population genetics.