We conducted three replicated field experiments to test the population response of two ecologically-divergent wolf spider species (Hogna helluo and Pardosa milvina) to three correlates of landscape fragmentation: area reduction, spatial subdivision, and increased edge to core ratio. We selected these two species because they differ in vagility and habitat selectivity. Hogna helluo is relatively large, averse to disturbed substrata, and has poor colonization abilities. Conversely, Pardosa is small, vagile, and will use barren, disturbed areas. In a test for the effect of area reduction on populations of the two wolf spiders, we destroyed 0%, 20% or 80% of randomly selected habitat islands in replicated experimental landscapes. We found that population densities of Hogna declined significantly, even at the lowest level of area reduction (20%), and that there was an increase in numbers of Pardosa. In a test for the response to an increase in landscape subdivision, we created four levels of habitat fragmentation in replicate plots. We found a significant decline in Hogna populations with increasing fragmentation. Pardosa populations did not respond to the fragmentation. In the third experiment we kept landscape area and subdivision constant, but manipulated the edge-to-core ratio. We found that populations of Hogna declined sharply with increasing edge, and that populations of Pardosa did not respond. These two syntopic wolf spiders have distinctly different responses to landscape fragmentation.