Biological invasions threaten biodiversity and ecosystem function, with pronounced negative effects on islands in particular [1-3].
Genetic studies of invasive species can identify the adaptive potential
of invaders to deal with new environmental conditions  or help to predict evolutionary responses to management practices (e.g. pesticides, biological control agents) .
Population bottlenecks affect many invasive species because they
frequently experience founding effects that reduce genetic variability,
but paradoxically, invasive species still manage to successfully
establish and adapt to new environments .
However, the effects of bottlenecks may be countered by the occurrence
of multiple introductions, high reproductive rates, and subsequent
migration between locally bottlenecked populations that are genetically
For invasive arthropod parasites, these factors are inextricably
linked with the distribution, genetics, and behaviour of host species [8-10].
The recent integration of molecular ecology with parasitology has
provided a path for answering a number of questions concerning the
genetic structure of parasite populations, which can uncover a wealth
of information regarding ecological and evolutionary processes for
invasive parasites .
Highly variable multilocus genotypes are particularly suited to
analyses of non-equilibrium or bottlenecked populations because they
provide adequate variation for assessing recent gene flow and
identifying migrants .
The introduced fly, Philornis downsi, is an avian
ectoparasite that is considered to be a serious threat to the
persistence of endemic finch populations on the Galápagos Islands [12-14]. Recently, P. downsi was given the highest risk ranking affecting endemic fauna in the Galápagos archipelago . Other pathogens affecting Galápagos birds such as avian pox virus  and intestinal protozoans 
are of less concern, but may also cause high fitness impacts under
certain conditions. The fly was first formally identified from Darwin
finch nests in 1997 and has since been found on 11 of 13 major islands
in nests of 14 endemic species [12,13]. However, P. downsi colonised the islands at least 40 years ago, as the fly was identified recently from collections made in 1964 . The blood-feeding larvae of P. downsi are associated with 62–100% nestling mortality in Darwin's finches [12,14,17], as well as physiological costs  and reduced growth rates in nestlings . Little is known about the ecology and biology of Philornis flies and the dispersal behaviour and population genetics of the genus Philornis or of any other myiasis-causing parasite of birds [reviewed in ].
One potential control method to eradicate P. downsi is the
sterile insect technique (SIT), which is renowned for its effectiveness
at eradicating or suppressing fruit fly and screw-worm fly populations
across the globe [19,20].
SIT involves the large-scale release of laboratory-reared sterile male
(and/or female) flies that eventually suppress fly populations by
reducing population fecundity [reviewed in ].
SIT requires a thorough understanding of the reproductive ecology and
population dynamics of the target species. The effectiveness of SIT is
affected by the occurrence of genetically divergent 'strains' of the
target species across the geographic area under control because this is
detrimental to the mating success of sterile flies [19,21,22].
Specifically, high genetic divergence may reflect differences in
behaviour and/or morphological characteristics that result in mating
incompatibility among populations of the target species [21,23].
Thus, target populations that show low genetic divergence are not
likely to show reproductive isolation and influence the success of a
particular sterile strain.
The Galápagos archipelago offers a unique system to examine the
population genetics of an introduced avian parasite that causes severe
fitness costs and that is still within a relatively early phase of
invasion. We collected parasites in 2004, 2005 and 2006 from three
islands of the Galápagos. Using mitochondrial data, we firstly
determine whether the three island populations from which we sampled
are of the one fly species. We then use microsatellite data to examine
gene flow within and among islands to: (1) determine whether dispersal
and genetic divergence are occurring among islands and between habitats
within islands (wet highlands, arid lowlands), (2) determine the
presence of population bottlenecks resulting from the invasion process,
and (3) determine whether inter-island genetic differentiation may be
of concern to the potential success of an archipelago-wide SIT program
for controlling P. downsi.