As dominant herbivores and notorious pests in their native Neotropics, introduced leaf-cutting ants have the potential for ecological and economic harm. Although a large-scale invasion of leaf-cutting ants has not occurred, an isolated introduction in the Caribbean islands of Guadeloupe provides useful insight into the progress of such an invasion. Since being first detected in 1954, Acromyrmex octospinosus has colonized virtually all available land area, defying an aggressive control campaign and damaging agriculture. I attempted to reconstruct the origins and spread of the invasion, as well as screen for the presence of garden pathogens, which could be used for biological control. Mitochondrial sequencing of the A. octospinosus complex throughout the Caribbean showed that the probable source of the invasion lies on Trinidad and Tobago or northeast South America. Using historical records and field surveys, the invasion's rate of spread was estimated at 0.51 km/year. Microsatellite genotyping further confirmed the limited dispersal abilities of A. octospinosus, showing the presence of isolation by distance (even in a relatively small geographic area) and suggested ubiquitous local inbreeding. Although the invasion likely resulted from the introduction of a single colony, microsatellites showed a high level of genetic variation in the introduced population, likely as a consequence of multiple mating by the queen. A survey showed that the specialized fungus garden pathogen Escovopsis exists on the islands, suggesting that the successful spread of the ants was not due to escape from this parasite. Given that chemical control has failed in the past and that biological control using specialized garden pathogens seems improbable, only vigorous quarantine and inspection programs may prevent wide-scale leaf-cutting ant invasions in the future.