The historian A. C. Crombie identified six styles of reasoning in the history of the sciences. Crombie's idea, motivated by history, suggested to me a philosophical programme for thinking anew about Scientific Reason. My interest in the present paper is not to defend the older parts of the styles of reasoning project, but to develop a further aspect of the programme. In particular, this paper turns to the fifth style in Crombie's list, probable reasoning (or statistical inquiry). I shall show that one of the novelties connected with probable reasoning, is a new kind of object, the population. At the end of this paper I shall suggest many a novelty that accompanied the evolution of probability and statistics as a way of reasoning and finding out.