L'étude des activités hydrologiques menées au cours des cinquante dernières années dans les pays francophones de l'Afrique sub-saharienne apporte un précieux éclairage sur les difficultés rencontrées, les modes de collaboration adoptés et l'ampleur des défis à relever dans le domaine de l'eau à l'aube du troisième millénaire.Ceci est l'occasion de tracer les étapes marquantes de l'acquisition de la " connaissance hydrologique " depuis l'ère pionnière des années 50 jusqu'aux programmes scientifiques les plus récents faisant appel à des technologies avancées. Cette rétrospective permet également d'apprécier l'évolution des modes de partenariat entre acteurs scientifiques, techniques et économiques du Nord et du Sud.Ces travaux et recherches ainsi entrepris et développés montrent à la fois l'importance du savoir acquis et la nécessité de poursuivre les études en cours. Ils dressent aussi un constat encourageant quant à l'avenir des projets régionaux africains en hydrologie grâce au renforcement des relations scientifiques entre pays de la région., This article describes a study of hydrological research carried out in the French-speaking countries of sub-Saharan Africa over the past fifty years, by all the parties concerned.The work sheds useful light on the difficulties encountered, the types of co-operation adopted, and the scale of the challenge now facing the region in the matter of water resources, on the eve of the third millennium.The first stage in the acquisition of hydrological knowledge in the region was the pioneer period of the 1950s, when many different kinds of water demand were emerging very fast but hydrological information was almost nil. Hydrometric networks incorporating 1,500 instrument stations had to be rapidly set up for an overall survey of water resources. 200 representative or experimental catchments were equipped. The data obtained from the stations provided practical answers, meeting the operational needs of development, though in many cases the data acquired were used again, later on, for pure research purposes. The data gathering was a huge task, carried out on a range of different geographical scales. To achieve it, hydrologists had to develop appropriate methodologies for the region's tough climate and difficult field conditions; they also had to train highly motivated, seasoned hydrologists.During the sixties, the first hydrological monographs on the major river basins were completed,, rue Lafayette - research was conducted on regimes and ten-year flood peaks, and findings on the small catchments were collated. All this provided a first sketch of the region's main hydrological features.Two major upheavals marked the end of the 1960s: a long, disastrous drought and the information technology revolution.The scale of the hydropluviometric deficits recorded over several decades, amounting in some catchments to 30%-40%, led scientists to wonder how stable water resources really are and whether the notion of "normal" in hydrology or rainfall has any validity. The question was especially relevant because the design of all major African water engineering schemes completed around 1960-65 was necessarily influenced by the flow rates recorded in the previous, wet period. Changes in ground surface states and the environment in general, under the impact of climate change and increasing human activity respectively, made a more multidisciplinary approach to the hydrological cycle indispensable. Information technology also had a major impact, as the many numerical data gathered could at last be properly analysed and exploited. User-accessible reference databases were developed and processing software packages like Hydrom and Pluviom were widely distributed. It was now possible to model the relations between rainfall and discharges at a detailed scale; modelling improved steadily, first with the introduction of fine-mesh models, then of coupled surface water-groundwater models.As regards water engineering work, flood routing models were significantly improved for the Niger and Senegal rivers, and dam operation simulation models made spectacular strides. Of course, data processing greatly facilitated the use of statistical laws for everything to do with engineering hydrology.For several decades, metrology made little progress; but the instruments in use were robust and easy to use, and so well suited to local conditions. Not until the early 1980s was there significant technological progress in sensor technology; meanwhile civilian satellites made their debut and were used from the outset to facilitate hydrological data gathering, verification and transmission. Two applications illustrate this use locally. One is the Hydroniger project, involving eight countries bordering on the Niger and designed to produce an operational, real-time hydrological forecasting system on the river basin; the other is the Onchocerciasis Control Program, with WHO. In this program, discharge data in only slightly delayed time has made it possible to calculate the right dose of insecticide to add to infected rivers. Over the past ten or fifteen years, hydrological research has considerably extended its scope, to include spatialised hydrological parameters, soil/water/plant/atmosphere relations, soil erosion and conservation, geochemistry, hydrochemistry, etc. Meanwhile, sub-Saharan Africa has not been left out of the international organisations' programs involving operational or pure research hydrology, and it is making its contribution to the great challenges that face us with regard to water resources and management. Examples are the WMO WHYCOS project to establish permanent monitoring systems; the AOC-FRIEND program, a regional research project, which has been growing in strength; and the Hapex-Sahel experiment, designed to provide more information on the thermal and moisture characteristics of soil and vegetation, for integration into general atmospheric circulation models.Forms of collaboration among the partners involved in hydrology in this part of Africa have naturally changed over the course of fifty years. In the beginning, demand from local technical services set up a particular form of partnership with hydrology experts from the North. After independence, scientific and technical co-operation developed along with a policy for training management-level staff to build up national hydrology services. In many countries, however, the national services only gradually took full charge of their hydrometric networks, with technical assistance continuing for a long time. During the seventies and eighties, most countries acquired scientific or technical infrastructures that played host to hydrologists from the North for joint research programs and studies, with gradually increasing involvement by African water research scientists.After long years of work gathering, interpreting and applying data, the hydrological characteristics of this part of Africa have been established and the groundwork for a sound scientific partnership between Northern and Southern partners has been laid. Many of the region's water problems are linked to the difficulty of meeting constantly rising demand from fluctuating water resources; unrelenting effort is needed to resolve these problems.In this regard, there are promising signs for the new millennium: the growing community of African hydrologists is increasingly active in major international programs; links have been forged among those involved in Africa's various regional programs; and there are now thriving scientific associations like the Association of African Hydrologists.