Ecological biogeography of arthropods on Spartina islands in northwest Florida

Experimental tests of the equilibrium theory of insular biogeography with the terrestrial arthropod fauna of small islands in Northwest Florida are described. Six islets composed of pure Spartina alterniflora (area: 56 m^2 to 1023 m^2; distance from mainland: 29 m to 1752 m) were defaunated with methyl bromide gas after intensive, inclusive censuses of their terrestrial arthropod inhabitants. Two additional islands and a large S. alterniflora stand on the mainland were not defaunated and served as controls. The recolonization process was monitored by weekly nonlethal censuses of all the sites. Four large cages were placed over other, similar, islands to determine the persistence of populations trapped inside the cages in the absence of immigration. In addition, six traps located in the study area showed the general movement patterns of arthropods. A distinction was made between transient and resident species colonizing the islands based partly upon 3 previous yr of data on the arthropod communities of the area. Recolonization of the defaunated islands was rapid. Allowing for seasonal variations typical of the area, the defaunated islands attained a species richness corresponding to the predefaunation levels in approximately 20 wk and oscillated about that level for the rest of the 53-wk experimental period. Extinction probabilities were inversely correlated with area, and immigration rates during the first 20 wk after defaunation were positively correlated with area. Extinctions and immigrations occurred at low levels following week 20, indicating turnover at approximately 0.14 species/wk. No statistically significant effect of isolation on island immigration rates could be demonstrated. Numbers of species collected on the traps were significantly affected by distance from the nearest large land mass and trap orientation. Only area proved to be significantly correlated with insular species numbers.