CONICET | Buscador de Institutos y Recursos Humanos

Abstract Plant communities developing in aban- doned semi-natural areas are being increasingly dom- inated by invasive exotic species. How these novel residents affect re-colonisation by native perennial species, a process generally assumed to be seed limited, remains little explored.We examined the relative roles of dominant exotic grasses and seed dispersal in limiting the richness and abundance of native perennial grasses in an old-ﬁeld grassland community. We also tested whether native grass recruitment depended on the identity of resident exotic species. A seed addition, single-pulse removal experiment was established in a 20 year old ﬁeld in the Inland Pampa of Argentina. Seeds of seven native perennial grasses from a nearby relict grassland were sown into intact and disturbed patches dominated by one of four exotic grasses. Species richness and biomass were measured after 2 years from sowing. Seed addition alone had little effect on native grass richness or biomass, with only one sown grass establishing in intact patches. Native grasses successfully colonised disturbed patches dominated by the exotics Loliummultiﬂorum,Cynodon dactylon or Sorghum halepense. In contrast, patches dominated by Festuca arundinacea repelled sown native grasses, regardless of disturbance treatment. Seed addition increased total plant richness in both disturbed and intact patches but did not affect total aboveground biomass. Our results show that recovery of native grasses during old-ﬁeld succession is hierar- chically constrained by seed arrival and site pre- emption by exotic grasses. Thus, re-establishment of native grass assemblages may only occur at the expense of displacing exotic resident plants. This highlights the importance of niche-limited species assembly in novel, native/exotic plant communities.