Dispersal of wet land plant seeds by mallards: Influence of gut passage on recovery, retention, and germination

Abstract

Seed dispersal is an important process for plants, but may be particularly important for plants occurring in spatially isolated habitats like wetlands. Variation in the geographic distribution of wetland plant species may be strongly influenced by their ability to use waterbirds, particularly waterfowl to enhance dispersal. We used controlled feeding and germination experiments to investigate recovery, retention time, and germination for the seeds of 10 wetland plant species (Chenopodium album, Digitaria ischaemum, Echinochloa colonum, Echinochloa crusgalli, Eleocharis palustris, Panicum dichotomiflorum, Polygonum lapathifolium, Polygonum pensylvanicum, Rumex crispus, and Scheonoplectus maritimus) and compared how these metrics varied with seed characteristics. We fed a known number of seeds to captive raised mallards (Anas platyrhynchos) and collected fecal samples every 4 hours for 48 hours; all recovered seeds were planted in seedling trays and watched for 60 days to monitor germination. We conducted 10 trials and fed each seed species in each trial, and included germination controls of non-consumed seeds. Overall, 19.0 ± 1.8% (mean ± SE) of seeds were recovered and of recovered seeds 7.6 ± 1.2% germinated. Recovery ranged from 1.9 ± 0.6% in D. ischaemum to 51.0 ± 4.7% in S. maritimus and germination ranged from < 0.5% in D. ischaemum to 28.5 ± 5.7% in R. crispus. Recovery and germination were not related to seed size or mass (p > 0.5) but recovery was positively correlated with seed fiber content (r2 = 0.44, p = 0.04). Control seeds germinated better than fed seeds for all species except S. maritimus, where fed seeds germinated better. Germination percentage declined with gut retention time for four of seven species. We suggest despite the large differences in viable seeds recovered, mallards may be important dispersers for the seeds of most species we studied and hypothesize that observed variation may be attributable to different plant strategies that relate to reproductive tactics and habitat type. © 2008, The Society of Wetland Scientists.