The cultivation of common reed (Phragmites australis) is one of the most promising practices of paludiculture on fen peatlands. This highly productive grass has a high adaptation capacity via high levels of genetic diversity and phenotypic plasticity. In this study, a reed experimental site established on a degraded fen in 1996/97 with a mixture of monoclonally (meristematically propagated plantlets) and polyclonally (seedlings) planted plots was investigated by microsatellite genotyping. All of the nine genotypes of the monoclonal planted plots were recovered and could be genetically characterized; invasion by other genotypes was negligible. Similarly, the polyclonal plots remained in this state, no prevalence of a single genotype was found. The growth characteristics of the five quantitatively investigated genotypes clearly differed from each other: dry biomass per stem 5–18 g, panicles per m2 20–60, average stem diameter 3.5–6 mm, height 170–250 cm. Similarly, the persistence (dominance at the planted plots) and invasiveness (ability to invade neighboured plots) of the genotypes were different. These results show that stands of reed are extremely persistent even if established with genotypes that are likely not to be locally adapted. Their genetic structure remained stable for at least 24 years, and this is largely independent of planting density (1, 4, and 10 plants per m2). Our results indicate that farmers may be able to maintain favourable genotypes for many years, thus the selection and breeding of reed as a versatile crop for rewetted peatlands is a promising objective for paludiculture research.