Diabetes mellitus, Banting and the discovery of insulin: a paradigm of bibliographical serendipity