Historical demography did not appear from out of nowhere. Before its success in 1958, Louis Henry was already working in a promising field, in which several model co-existed or competed with one another. During the early past-war years studies undertaken by historians, demographers, and geographers were based on various types of past censuses. The idea of reconstructing statistical series from data in perish registers only began to catch on and prevail during the nineteen-fifties. The intense debates centred on periodization provide an illustration of the way in which this idea evolved. They show that in spite of pro grammes which were often promising and logical, Louis Henry's technical rigour would not by itself have been sufficient to win historians over to his views : he was successful in imposing his method and thus gave an impulse to the development of historical demography, because he was able to include in his own projects, both problems defined by various protagonists as well as the opposing views of the historians.