Many studies have examined seed removal by rodents without regard to possible biases from inadvertent olfactory cues left by the researcher. I compared removal rates of seeds that had direct contact with human skin (scent treatment) with those that had no direct contact (no-scent treatment). Seeds of two species were alternated at 5 m intervals along a straight transect in a native grassland in northwestern Illinois. Each treatment had 100 seeds, equally divided among two woodland species that are regularly dispersed by birds into grasslands. Seeds were monitored for removal for 30 d. The rate of seed removal was significantly faster for the scent than no-scent treatment. At the end of the study 88% of scented seeds had been removed compared with 55% of the no-scent seeds. Researchers should minimize unintentional olfactory cues during seed removal studies in order to get more accurate estimates of the impact of seed removers.