Weed seeds present in harvested silage have to survive silage fermentation and rumen digestion before they are dispersed as a contaminant of manure. Therefore, producing crops that are ensiled could lower the seed dispersal of weed escapes. This study is aimed at evaluating the viability of seven weed species after storage in experimental mini-silos filled with corn or alfalfa. Nylon mesh bags, each containing one hundred seeds of a weed species, were inserted at random locations in mini-silos filled with silage corn or alfalfa and stored for one, three or six months. The experiment included five mini-silos per storage time as well as untreated seeds. Water imbibition by intact seeds was also evaluated to determine if it could be related with survival in silage. After three and six months of storage few seeds were viable in any treatment (<0.1% of all seeds tested). Differences between weed species and silage type were observable after one month of storage and could not be related to seed coat permeability as measured by water imbibition. Ensiling for three to six months, or more, could be used to kill harvested weed seeds. Further evaluations in commercial farm silos could be done to support results.