Over the past 20 years there has been a growing recognition that land develop? m nt plans in the tropics should be ased on surveys of the physical environ? ment, or natural resource surveys. The immediate aims of these surveys have been of a local nature: to enable development to make the optimum use of resources and to avoid environmental hazards, particularly those consequent on the disturbance of the ecological balance. But the maps and monographs produced for these applied pur? poses have also contributed substantially to knowledge of the nature of the physical environment of the tropics. Each group of natural scientists concerned with resource surveys realized the need to provide a framework to which detailed local surveys could be related. Nowhere was this more necessary than in Africa, where scientists from Britain, France and Belgium have tended to work independently of each other, establishing different concepts and methods, particularly with regard to environmental classification systems. Inter? national coordinating organizations were established, and under their auspices a series of maps of environmental factors for the African continent have been produced. These include two vegetation maps of a scale of i: 10 million, one showing physiognomic vegetation types (Keay, 1959) and the other floristically-based grassland types (Rattray, 1960; previously reviewed in Geogr. J. 128 (1962) 527); two climatie atlases, with the climatie elements mapped at scales of 1:5 million to 1:30 million (Jackson, 1961; Thompson, 1965, previously reviewed in Geogr. J. 132 (1966) 291-2); and two maps at 1:5 million, which it is proposed to discuss here, of geology and of soils (Furon and Lombard, 1964; D'Hoore, 1964). It has now become possible to abstract a range of environmental information, classified on a uniform basis, for any selected location in Africa (south of the Tropic of Cancer). It would be naive to expect a high level of accuracy for these maps, large parts of which are based on reconnaissance surveys at small scales or interpolations between scattered data; but at the least they summarize the views of the appropriate specialists, with all available information at their disposal. The small-scale maps of African geology, climate, vegetation and soils carried by atlases may now be re-drawn on a comparatively well-established foundation. The maps of geology and soils are both cartographically well-produced. The Geological map covers the continent in eight non-overlapping sheets; a ninth gives the full legend, and the whole can be assembled as a display map some six feet square. Parallels and meridians are shown at 50 intervals, with i? intersections marked by small crosses; the latter in no way obscure the geological detail, yet permit any chosen area or point to be located rapidly. Geological information is shown by colour-shading and symbols, although there are some discrepancies between the combinations of these ap? pearing on the map and in the legend. The Soil map comprises seven sheets, with parallels and meridians shown at 20 intervals. A substantial overlap between sheets renders it more convenient for reference. On both maps the relevant parts of the legends are shown on each sheet, but in the case of soils the full legend is found only in the explanatory monograph. The Geological map is accompanied by a list of source maps for each territory, the scales of which provide some indication of the intensity of survey and hence the accuracy to be expected. For the Soil map there is only a list of contributors, many of whom who are no longer working in their respective countries. Courtesy and politics forbid the inclusion of an inset map showing relative accuracy on international compilations of this kind. The nature of the distribution pattern of soils is such that profiles with widely different properties, separated at a high level in any systematic classification, occur within small areas. Consequently in mapping on a continental scale, composite units must be employed. The 63 soil types contained in the classification scheme are termed ^ Dr. Anthony Young is Lecturer in Geography at the University of Sussex, and has worked as Soil Surveyor in Malawi. Earlier this year he was geomorphologist on the Royal Society/R.G.S. Mato Grosso expedition.