Genetic structure of blow fly (Diptera: Calliphoridae) populations has remained elusive regardless of high relatedness within wild-caught samples. The aims of this research were to determine if the implementation of a high-resolution spatiotemporal sampling design would reveal latent genetic structure among blow fly populations and to elucidate environmental impacts on structure. Adult female black blow flies, Phormia regina (Meigen), were collected from nine urban parks in Indiana, USA over three years. Fly gut DNA was genotyped at six microsatellite loci, with subsequent amplification and sequencing of vertebrate mitochondrial DNA from the same source. Flies were also screened for vertebrate fecal metabolites. Latent clustering revealed four genetic groups which were interpreted as 11 distinct temporal populations. An analysis of molecular variance of temporal populations revealed stronger genetic differentiation (F_ST = 0.048, F’_ST = 0.664) relative to geographic populations (F_ST = 0.009, F’_ST = 0.241). Mean kinship within temporal populations was higher than expected in a panmictic population (R = 0.032 ± 0.088). Weather conditions (i.e., wind speed, precipitation, humidity, temperature) and vertebrate resource availability in the local environment significantly impacted the observed genetics of P. regina. Twenty-five vertebrate species were detected from flies, and 16% of flies collected in 2016 – 2017 tested positive for vertebrate feces, suggesting many varied resources are important for maintaining high gene flow among geographic populations. A complex interplay between biotic and abiotic factors, as well as the flies’ own extensive dispersal abilities, seems to drive the strong temporal structure of this species in the Midwestern US.

Blow Fly Collections

Spatiotemporal blow fly collections were made over a three-year period (2015 – 2017) in Central Indiana, USA. Blow flies were sampled from nine geographic sites over 36 sampling periods. Sampling periods occurred from June to September in 2015 and from March to October in 2016 and 2017. Collection sites consisted of urban parks and other public areas in four different counties (Table 1).  Fly collections were made with an aerial sweep net and an aged chicken liver bait inside a mesh-screened container to prevent flies from physically landing on the bait without hindering their ability to detect the smell of the bait [86]. Fly sampling lasted 20 min starting at the time of bait exposure. Temperature and relative humidity were measured throughout the collections using a Datalogging RH/Temperature Pen (SPER Scientific, Scottsdale, Arizona, USA) elevated ~1 m above the ground. Wind speed and direction were also measured at three arbitrary times during each collection using a Digi-Sense® Mini Vane Anemometer (Cole-Parmer, Vernon Hills, Illinois, USA). After the 20 min collection window, flies were immediately killed in 70% ethanol for transportation to the lab and stored at -20˚C. Flies were identified to species using a dichotomous key [87] and fly richness and evenness per sample were calculated using the vegan package [88] in R Studio v1.4.1717 [89].

Dissections and DNA Extractions

Gut dissections from adults of the black blow fly, Phormia regina (Meigen), were performed according to previously published protocols [90] on a maximum of 10 randomly chosen females from each sample. A standard phenol:chloroform:isoamyl alcohol DNA extraction method was performed on each individual gut dissection, with a final elution in 35 uL of TE buffer. This procedure allowed for the simultaneous extraction of fly DNA (from the gut epithelial tissue) and vertebrate DNA (from the last meal within the fly gut). The normally discarded phenol layer of the DNA extraction was also screened for fecal metabolites [90].

Genotyping

Six polymorphic microsatellite loci were used to assess population structure [24]. Amplification of each sample was carried out in two separate multiplex reactions each containing three primer pairs. One reaction contained primers L3, L9, and L12, and the other contained primers L8, L13, and L14. Each reaction consisted of 5.0 μL 2x PCR MasterMix (Promega), 0.5 μL 1x Bovine Serum Albumin (BSA), 0.5 μL nuclease-free water, 1 μL of each primer, and ~1ng DNA. Concentrations for each primer were as follows: 2.5 μM L3, 1.5 μM L9, 5.0 μM L12, 2.5 μM L8, 2.5 μM L13, and 2.5 μM L14. PCR conditions consisted of an initial 2 min denaturation step at 95˚C, nine touchdown cycles of a 15 s at 95˚C, 30 s from 63˚C to 55˚C, and 72˚C for 2 min, followed by 22 cycles of 95˚C for 15 s, 55˚C for 30 s, and 72˚C for 2 min, with a final extension at 72˚C for 7 min (Mastercycler Pro thermocycler, Eppendorf®, Hamburg, Germany). Following amplification, 1 μL of each PCR product was added to 9.7 μL HiDi formamide (Thermo Fisher Scientific, Waltham, MA) and 0.3 μL LIZ internal size standard (Thermo Fisher Scientific) for fragment analysis on a 3500 genetic analyzer (Thermo Fisher Scientific) and alleles were scored using GeneMarker® software (SoftGenetics®, State College, PA, USA).