Blow and flesh flies (families Calliphoridae and Sarcophagidae) feed on carrion, open wounds of living animals and/or faecal matter and generally oviposit on the first two substrates. Weird as it might sound this makes them perhaps ideal candidates for an alternative way of sampling DNA from wild mammals.
A group of researchers from Germany demonstrated that mammal monitoring through carrion flies is possible and may even overcome some of the impediments associated with the use of blood eating insects and leeches. Ttsetse flies and leeches live in restricted habitats and many mosquito and tsetse fly species are host specific. Both problems don not occur with blow and flesh flies. As we all know they are usually not picky when it comes to habitat or host choice.
The team captured several flies that emerged from carcasses found in two forests —the Tai National Park in Côte d’Ivoire and the Kirindy Forest in Madagascar—and extracted the DNA in their guts to get a snapshot of the mammals in each area. The random collection of 75 and 40 flies at the two parks allowed them to recover of DNA from 16 and 4 mammal species, respectively. For Kirindy, this corresponds to 13% of the known local mammalian community (31 species). The mammal biodiversity of the Tai National Park is not yet fully characterized but striking outcomes were the sampling of six of the nine species of the local primate community and the detection of Jentink's duiker (Cephalophus jentinki), a very rare and endangered antelope species whose entire population is estimated to be <3500 individuals world wide.
Carrion fly DNA analysis might offer a broad range of potential applications linked with the assessment and monitoring of mammalian biodiversity. Ed Yong calls it census-taking via corpse-eating.
The only critical comment I have is my usual "standards lament". The colleagues used 16S and 12S and I couldn't find any explanation in the paper why they didn't use COI which would have allowed them to tap into a much larger library and allowed us to compare their data to other barcoding studies.
h/t Bob Hanner