Published online by Cambridge University Press: 20 January 2017
Natural dissemination of johnsongrass seeds as well as the effect of combineharvesting on this process were studied in corn fields. The estimation ofnatural dispersal was carried out by two different methods, collecting seedsthroughout the season using seed traps and sampling soil–surface seedabundance before harvest using a vacuum device. Both methods showed the samedispersal pattern. A minimum of 84.6% was dispersed in the first 2 m fromthe focus and a maximum of 1.6% was dispersed beyond the first 5 m. Anaverage of 76.3% of these dispersed seeds were lost or buried after sheddingbut before harvest. Seed dispersal by the combine harvester was estimatedfrom the difference between soil–surface seed abundance in the same sitespre and postharvest. Although the quantity of seeds dispersed by the combinewas similar to those dispersed by natural factors, dispersal distances weresignificantly higher. Around 90% of the dispersed seeds were found in thefirst 5 m forward and backward of the combine direction from the infestationsource, and 1.6% of the seeds were found beyond 22 m forward and 10 mbackward of the combine direction from the infestation source. A largeproportion of the seeds dispersed were dormant or not viable. It isconcluded that the major role of sexual reproduction in johnsongrasspopulation dynamics may be to spread the risks, promoting dispersal in timeand space.