POPULATION-STRUCTURE OF THE RARE, LONG-LIVED PERENNIAL GENTIANA-PNEUMONANTHE IN RELATION TO VEGETATION AND MANAGEMENT IN THE NETHERLANDS

1. After analysis of population age-state spectra of Gentiana pneumonanthe in a variety of vegetation types, three different population types could be distinguished: (a) 'invasive' or 'dynamic' populations, characterized by high densities of seedlings and juveniles relative to the adult age states, (b) 'normal' or 'stable' populations with adult age states prevailing, but with low densities of seedlings and juveniles, and (c) 'regressive' or 'senile' populations, consisting only of adult flowering and adult vegetative individuals. 2. The structure of the surrounding vegetation appears to be very important in determining the population structure of the marsh gentian. In particular the percentage of bare soil surface and the cover of the litter layer explain most of the variation in the density and proportion of different age states in the populations. The area of bare soil surface in the vegetation increases the percentage of seedlings and juveniles in the population and, consequently, decreases the proportion of generative adults, while the reverse is true for the cover of the litter layer. 3. In heathlands, a clear trend can be observed in the age state structure from 'invasive' populations in young successional stages, via 'normal' populations in relatively stable situations, to 'regressive' populations in late successional stages or unmanaged areas dominated by grasses or shrubs. In hay meadows, 'normal' populations are usually observed, except where mowing has been stopped or where a combination of early mowing and acidification exists. This leads to closure of the community by grasses or bryophytes, respectively, and therefore to a 'regressive' population structure. 4. Individuals in heathland and grassland populations seem to have a different life-strategy because of the differences in vegetation dynamics between these two habitat types. Ecotypic differentiation in life-history characteristics may therefore be expected. 5. It is argued that, as a basis for conservation and management, monitoring the age state structure of populations of rare perennial plant species is far more effective than counting only the flowering individuals, as is mostly done.