In a secondary succession series in grasslands, which is induced by a drastic change in management regime, the annual hemi-parasite Rhinanthus angustifolius :* appears, builds up a population (phase I) and becomes temporarily abundant (phase II) whereupon the population decreases to a low level (phase III), but with considerable fluctuations. Different vegetational succession stages (corresponding to the three phases of population development) were found in closely adjacent lots which had been managed for different periods of lime. In this paper results of four years demographic work on Rhinanthus angustifolius in these three succession stages are analysed. The shown differences in population phases cannot be explained by selection or density dependent behaviour of the plants. The biggest loss factor in the population is in the seed stage in all phases. No clear correlation exists between seed production and number of seedlings the following year. Seedling survival on average is high in phase I but can also be quite high in the other phases. Seed capsule production early in phase I is much higher than in phase II and III, and no difference in seed capsule production occurs between phase II and III. Predation of the seeds in the seed capsules is highest in phase II and can reach 30%. When the seeds are shed seed predation is high just after the shedding, but much seeds disappear in spring when germination begins. Differences between the phases are clear but the year to year fluctuations are of the same order of magnitude or even greater. The spatial distribution of a colonizing population of Rhinanthus angustifolius was mapped for 4 years, numbers increase continuously despite the yearly fluctuations. Local environmental conditions are mainly responsible for differences in densities of Rhinanthus angustifolius between the succession stages. Fluctuations in numbers are caused by climatological factors and local conditions (e.g. species composition and microclimate).