A manipulative experiment was done in corn fields and their adjacent forests using enclosures that restricted access to ground-dwelling spiders. Enclosures were either closed from the adjacent habitat but open to ballooning and ground-dwelling spiders (using holes cut in the side of enclosures) or were open plots (controls). This allowed us to test the role of ballooning compared to cursorial dispersal of ground-dwelling spiders within these habitats. A reciprocal substrate treatment was included in which leaf-litter was added to cornfields and removed from forests to test the interaction between mode of dispersal and habitat use. Ninety species were collected using visual surveys and with pitfall traps. More species were collected in cornfields, and more individuals were collected in litter-addition plots, but we uncovered no interaction between substrate treatment and enclosure type. However, enclosures that excluded cursorial spiders had fewer mature and immature spiders, suggesting that cursorial activity (at a small spatial scale) is an important mode of dispersal within both types of habitats.