Longevity of seeds of Poa pratensis and Lolium perenne as affected by simulated soil tillage practices and its implications for contamination of herbage seed crops

Abstract In grass‐seed production, the purity of the product is of major importance and high purity can only be obtained by preventing contamination with other species and cultivars during the whole crop rotation. In this context, seeds of previously grown grass species or varieties are particularly important. This study investigated the influence of different soil tillage practices on persistence of newly shed seeds of two grass species, Poa pratensis and Lolium perenne. Leaving the seeds at or close to the soil surface reduced seed survival of both species to low levels compared with strategies that involved a deep incorporation of the seeds shortly after seed‐shedding. Survival rate of seeds of P. pratensis after 1 year was reduced to a low level when seeds were left at the soil surface for 2–3 months before incorporation. Seeds of L. perenne were less persistent and survival rate after 1 year was generally very low if the seeds were left near the soil surface for approximately 1 month before deeper incorporation. Use of the stale seedbed technique to reduce the seedbank of the two species in the autumn ahead of a spring‐established grass‐seed crop was also found to be effective in reducing the potential risk of contamination.