Published in 1939, this book was the first to develop a fundamental theory of scientific inference based on the ideas of Bayesian statistics. Recent advances in computer power and availability have brought Bayesian statistics into the limelight and make this book a must for all serious statisticians. Though mathematically very demanding, the principles and examples are clear enough, and Jeffreys' trenchant references to critics of Bayesian methods often made me laugh out loud - an unusual experience with an advanced statistics text. Its selection as an Oxford Classic Text is unusual but well judged.