This paper outlines the contours of a historical geography of science. It begins by arguing for the relevance of spatially oriented histories of scientific thought and practice. The paper then considers three different historical geographies of science: those concerned with the places and spaces of science, those that detail the spatial contexts of scientific endeavour, and those that analyse the internal ‘cartographies’ of scientific theories and methods. The paper concludes with a discussion of other possible avenues of investigation in this field. Scientific knowledge is made in a lot of different places. Does it matter where? Can the location of scientific endeavour make any difference to the conduct of science? And even more important, can it affect the content of science? In my view the answer to these questions is yes.