A key goal of conservation biology is to prevent the spread of exotic speci
es. Previous work on exotic invasion has two limitations: (1) the lack of a
spatially explicit approach and (2) a primary focus on the net effect of i
nvasion by examining invasive species already present in the community. We
address these limitations by focusing on the arrival of a potential invader
into a community and use a spatially explicit approach to quantify the flo
w of seeds from tbe surrounding landscape into the interior of a forest. We
hypothesize size that the structure of forest-edge vegetation influences h
ow the edge mediates seed flux. To test our hypothesis, we experimentally a
ltered vegetation structure within 20 m of the edge to create two edge trea
tments: thinned and intact. We quantified the flux of seeds moving into the
forest interior across the two treatments. We used seed traps randomly arr
ayed on transects from 5 to 50 m into the forest. More seeds crossed the th
inned treatment than crossed the intact treatment to reach the forest inter
ior. In addition, seeds that crossed the thinned treatment dispersed furthe
r into the forest than those that crossed the intact treatment. These resul
ts were consistent throughout the period of maximum autumn dispersal, inclu
ding periods before and after leaf drop. Our results show that the structur
e of vegetation on the edge interacts with the flux of wind-dispersed seed
across the edge. We demonstrated that an edge with intact vegetation can fu
nction as a physical barrier to seed dispersal. Therefore, the structure of
vegetation on edges can influence the function of edges as barriers to see
d flux into the forest interior.