The history of medical education in Britain

quite so different as readers of Pour une histoire des sciences might suppose. In his well-known paper on the role of Cuvier in the history of science, Foucault also sought to show the intellectual conditions necessary for the emergence of (Darwinian) transformism, with Cuvier the pivotal figure "making possible" the reconstruction of eighteenth-century natural history so as to yield nineteenth-century transformism. It is true, of course, that Foucault and Roger told quite different stories, with different actors in the drama. Yet in Foucault's "Cuvier paper" it seems to me that he was doing something of which Roger might have approved, in principle at least, though the theoretical goals of the two scholars were not the same. So perhaps the debate was more about form than substance-as it now appears in retrospect. But let me return to the question of lexicographic anachronism. I do not like it. Take the question of histories of the earth, for example. These began to emerge as the characteristic way of "doing geology" in the early nineteenth century, as stratigraphy assumed its modern form, working with fossils, and piecing together a history of the earth's strata (and hence of the earth as a whole to some extent) on the basis of the study of rocks and strata in the field, somewhat as the historian works with the fragmentary documents that may be dug out of the archives. Now as I see it, it is an entirely worthwhile exercise to endeavour to examine, as Roger did, the seventeenth-and eighteenth-century theories of the earth, establish what their general features were, and how they were gradually changed so as to yield nineteenth-century stratigraphic geology. But if the word "history" meant something radically different to those thinking about the earth in the nineteenth century and in previous times, I think it advisable to make the distinction in the historian's vocabulary. For example, I have suggested elsewhere that it is helpful to distinguish between histories of the earth and the earlier genetic accounts of the earth's past, which might, for example, be of a "preformationist" or quasi-Leibnizian character. However, Roger's lexicographic anachronism would not require such a distinction; or perhaps he would not have found it worthwhile. Be that as it may, and whatever one may think about these rather general issues, there can be no doubt that historians will welcome this collection of Roger's work, some of it (particularly that …