We collected snakes from a site located in Volusia County, Florida which has a high prevalence of R. orientalis (Farrell et al., 2019). We retained snakes in the Reptile Lab at Stetson University for ~1 week to collect fresh fecal samples to detect patent R. orientalis infections via fecal wet mounts. We returned apparently uninfected specimens to their site of capture, and retained four infected snakes, including a black racer (Coluber constrictor, caught 7 September 2023), two banded water snakes (Nerodia fasciata, caught 26 August and 21 May 2023), and a ribbon snake (Thamnophis saurita, caught 10 April 2023). We confirmed these four snakes with pentastome eggs in their feces were infected by R. orientalis by (1) euthanizing, dissecting, and morphologically verifying the pentastome species after the end of the experiment (three snakes), or (2) examining an adult female pentastome expectorated during the experiment (one snake). Adult R. orientalis are readily distinguished from the native pentastome species infecting snakes in North America by head morphology (de Luna et al., 2022).

We housed the snakes individually in glass terraria with heating pads, newspaper substrate, water bowls, and two hiding locations. We fed the snakes once each week using frozen-thawed fish and rodents for N. fasciata, rodents and lizards for C. constrictor, and lizards for T. saurita. We checked terraria daily and collected fresh feces while avoiding urates and portions of feces comprised of mostly fur remnants. We collected 20 fecal samples from 26 September 2023 to 1 February 2024 (eight from C. constrictor, one from T. saurita, seven from N. fasciata #1, and four from N. fasciata #2). Immediately after collection, we mixed the feces with a sterilized (cleaned with 70% ethanol solution) metal spatula and subdivided each fecal sample into four similar-sized subsamples that we then weighed on an analytical balance. The fecal subsamples were small (mean mass=68.6 mg) to facilitate complete pentastome egg counts.

For each fecal sample, we randomly assigned one of four diagnostic procedures to the four subsamples. One subsample was used in its entirety to make a series of fecal smear wet mounts to allow us to count the total number of pentastome eggs per gram of fecal material. The wet mounts were prepared by mixing small amount of fecal material with approximately 0.1 mL of tap water on a glass slide. A cover slip was then placed over the mixture. Forceps were used to break apart any large, hard pieces of fecal matter, allowing them to disperse more evenly in the water for improved visualization on the slide. The other three subsamples were used with three flotation solutions, zinc sulfate (ZnSO₄, LabChem, specific gravity=1.18), sodium nitrate (NaNO₃, “FECA-Med”, Vedco Inc., specific gravity=1.26), and Sheather’s sugar solution (JorVet, Jorgensen Laboratories, specific gravity=1.27). In all flotation tests we placed a subsample in a fecalizer (BioRx Laboratories, Commerce, CA), which is a device consisting of a well and an insert with screening that grinds up the fecal sample and filters the debris from the eggs. One of the three test solutions was added to the fecalizer to the first fill line and the insert was rotated by hand to break up the feces. More solution was added to the fecalizer until the formation of a positive meniscus, at which point a glass cover slip was placed on top. The cover slide remained in position for ten minutes to allow the eggs to rise from the debris and accumulate on the cover slip, which was then carefully removed and placed on a slide. The slide was viewed under 50x brightfield microscopy and the number of pentastome eggs was recorded.

As these snakes were infected with pentastomes, we classified each test as a true positive result if any pentastome eggs were observed in a fecal subsample or as a false negative if no eggs were observed. To calculate the observed pentastome egg density in a fecal sample, we divided total number of eggs counted in the wet mount subsample by the wet mass of that subsample.

We checked snake cages daily from 30 December 2023 to 7 February 2024 and collected 18 fresh fecal samples (two from C. constrictor, six from T. saurita, six from N. fasciata #1, and four from N. fasciata #2). After we retrieved the feces, they were thoroughly homogenized with a spatula. Each fecal sample was separated into two subsamples of approximately equal size and then weighed (mean=89.3 mg, SD=4.9 mg). The subsamples were randomly assigned to either the fresh sample group (analyzed within three days of retrieval) or the aged sample group. The aged samples were kept in a desiccator with relative humidity of zero held at approximately 20°C and were held for an average of 16.5 days (range=14–21 days) before flotation. We used sodium nitrate flotation only (methods as described above) and counted the number of pentastome eggs observed. We divided the egg count by the subsample mass to get an observed egg density for each sample. Samples were classified as either true positive (at least one pentastome egg observed after flotation) or a false negative (no eggs observed).

The R. orientalis samples used for genetic analysis were either opportunistically collected from live reptiles and carcasses and shared via the Snake Lungworm Alliance and Monitoring (SLAM) program (https://invasionscience.ufl.edu/slam/) or collected through other studies. Sample sources included frozen-thawed carcasses (i.e., roadkill), live captive-held lizards and frogs (for other studies), and live animals handled for a multitude of other research projects. In total, we tested 611 samples from 27 snake species, four lizard species, and one toad species ( Supplementary File 1 ). These samples were comprised of 412 cloacal swabs and 98 fecal samples from snakes, nine R. orientalis larvae, and 43 R. orientalis adults. Additionally, other pentastome species included 49 adults of two native Florida species, Porocephalus crotali and Kiricephalus coarctatus, collected from snakes, as well as adult Raillietiella females from Hemidactylus spp. These Raillietiella specimens were morphologically consistent with blunt-tipped Raillietiella species (R. orientalis is sharp-tipped), specifically R. indica and R. taegueselfi (de Luna et al., 2022). Their host preference was also consistent with these species and for brevity, we will refer to these specimens as blunt-tipped Raillietiella throughout the manuscript.

Our dataset included 170 R. orientalis-negative controls, 155 R. orientalis-positive controls, and 286 snakes of unknown infection status. The negative controls included 90 cloacal swabs from carcasses that were confirmed negative, 24 fecal samples with no eggs detected, and samples from adults of three other pentastome species: K. coarctatus (n=18), P. crotali (n=20), and blunt-tipped Raillietiella (n=11), which were identified via morphology (all individuals, n=49) and further validated by DNA sequencing (n=18). Species identification of adults relied primarily on body shape, head, and hook morphology and established life cycles. We used CO1 primers (Folmer et al., 1994) to confirm our morphological identification of three P. crotali individuals and 18S rRNA primers (Brookins et al., 2009) to confirm our morphological identification of four K. coarctatus, the three P. crotali, and 11 blunt-tipped Raillietiella ( Supplementary File 2 ).

Negative controls also included an additional 24 fecal samples that we determined were negative through fecal wet mount microscopy, and 97 cloacal swab samples from snakes and lizards examined post-mortem that had no adult R. orientalis in their lungs. Additionally, we extracted DNA from six fecal samples that underwent fecal flotation and kept them separate from our other samples for analysis, due to the potential presence of inhibitors in the flotation solutions (three with sodium nitrate and three with zinc sulfate).

For fecal samples, positive controls included 42 pygmy rattlesnake (Sistrurus miliarius) fecal samples in which we observed pentastome eggs and one carcass that had R. orientalis in the lung. We used Sistrurus miliarius (S. miliarius) fecal samples as positive controls because S. miliarius almost exclusively hosts R. orientalis adult (Miller et al., 2020; Farrell et al., 2019). We confirmed these S. miliarius positive samples via Sanger sequencing (see below). There are currently no known diagnostic features of pentastome eggs that enable species differentiation via our standard wet mount techniques; and therefore, 31 additional fecal samples from other snake host species with pentastome eggs microscopically detected were considered unknowns. For cloacal swabs, the group of positive controls included (1) swabs from snakes that were dissected and found with R. orientalis adults in the lungs, and (2) snakes used in experimental infection studies with confirmed R. orientalis infections via conventional PCR and/or dissections. The positive control pentastome individuals included 9 larval R. orientalis isolated from the viscera of lizards and frogs used in experimental infections, and 43 adult R. orientalis from dissected snakes that were identified morphologically and confirmed via Sanger sequencing.

For DNA extractions, we used Qiagen DNeasy Tissue kits and adapted the manufacturer’s instructions for each sample type (Qiagen, Hilden, Germany). Adult pentastomes were cut with bead-sterilized forceps and scissors, then the 70% ethanol preservative was evaporated off the cut sections. Larval pentastomes preserved in 70% ethanol were placed in 1.5 mL microcentrifuge tubes and ethanol was evaporated off. DNA was extracted from adult (partial) and larval tissue (whole) using the Qiagen DNeasy Tissue kit, following the manufacturer’s instructions (Qiagen, Hilden, Germany). For the cloacal swab extractions, we followed the manufacturer’s guidelines of placing samples in 20 µL proteinase K and 180 µL of buffer ATL for incubation at 56°C. We incubated the swabs for three to six hours and then proceeded to add 200 µL of 100% ethanol and 200 μL of buffer AL. We used bead-sterilized forceps and scissors to transfer the swabs tipped-end facing up, into the spin columns and then pipetted 600 µL of the liquid into the spin column. Following the initial spin at 8,000 RPM for one minute, we removed the swabs and followed the manufacturer’s guidelines for the remainder of the extractions. The fecal extractions followed the protocol from Ayana et al. (2019), which was developed for the extraction of DNA from helminth eggs in fecal samples. Specifically, we used the modified DNeasy Blood and Tissue Kit protocol without bead beating (Qiagen, Hilden, Germany). We measured the concentrations of all samples using a Nanodrop spectrophotometer, except for three positive fecal samples, one cloacal swab, and one P. crotali sample.

We initially tested cloacal swabs, fecal samples, and R. orientalis adults with 18S primers (Brookins et al., 2009) and CO1 primers (Folmer et al., 1994) as these were the most recent PCR assays available in the literature. Given the lack of morphological methods for distinguishing pentastome eggs, reliable molecular detection techniques for R. orientalis infections in fecal samples were needed. Because no previous studies had evaluated the effectiveness of cloacal swabs or fecal samples with these assays, we utilized positive control samples confirmed to be infected with R. orientalis via dissection to test these methodologies and determine whether a new assay was needed. Specifically, we ran the 18S assay with 30 adult R. orientalis identified morphologically, 14 fecal samples with pentastomid eggs identified through wet mounts (three were from S. miliarius and considered positive controls, the other 11 were considered unknown), and 16 cloacal swabs, including the four positive control snakes used for the fecal flotation methods. We ran the Folmer CO1 assay with six adult R. orientalis and 16 cloacal swab samples described above. Due to the limited utility of these primer sets (see Results) and because no primers existed that were specific to R. orientalis, we developed an R. orientalis-specific PCR assay.

We focused on CO1 because 18S data in GenBank revealed no SNPs segregating between R. orientalis and other pentastomes. We designed R. orientalis-specific CO1 primers based on available sequences in GenBank. Specifically, we retrieved all available CO1 sequences across pentastomid species in GenBank, including R. orientalis, R. indica (and junior synonym R. hebitihamata), R. mottae, Porocephalus crotali, Armillifer spp., Kiricephalus coarctatus, Linguatula spp., and Waddycephalus spp. We performed multiple sequence alignment on these sequences to identify regions that were identical among all R. orientalis samples but contained single nucleotide polymorphisms (SNPs) that were fixed between R. orientalis and all other pentastome species. We designed primers within regions that had the highest density of these SNPs, corresponding to a target region spanning 194 base pairs of the CO1 gene: forward primer ROCO1F=5′-GCCTTCTCCATATTACTCCTC-3; reverse primer ROCO1R=5′- CGTATGTTGATGATTGTGGTAGTG-3′. The primers were designed for specificity and efficiency, taking into consideration factors such as melting temperature and potential secondary structures.

We optimized the PCR protocol using temperature gradient cycling protocols with R. orientalis positive controls and negative controls (including the other pentastome species). For our 20 µL reactions, we used 1 µL of template DNA, 0.5 µl of each primer (ROCO1F, ROCO1R) at 10 µM concentration, 13.95 µL of molecular-grade water, 2 µL of NEB OneTaq Buffer (5x), 0.8 µL of dNTPs (10mM/dNTP), 0.8 µL of BSA (10%), 0.2 µL of DMSO (10mM), and 0.25 µL of NEB One Taq enzyme. We amplified our 20 µL mixture in a thermal cycler (Bio-Rad T100) with a 3-minute denaturation period at 94°C, 35 cycles of: denaturation at 94°C for 30 seconds, annealing at 56°C for 30 seconds, and extension at 72°C for 60 seconds, and a final extension at 72°C for 7 minutes. To visualize the amplified product, we performed gel electrophoresis using a 2% agarose gel, run at 80 volts for 45 minutes.

We Sanger sequenced 212 PCR products from our R. orientalis protocol that produced bands of the expected size (~200 bp), as well as 22 additional PCR products (7 fecal samples with unknown eggs, 13 unknown cloacal swabs, and three positive control cloacal swabs) that had nontarget bands (larger than expected or smeared). All sequencing was performed by Eurofins Genomics (Louisville, KY, USA).

For fecal flotation analysis, we used a 4×2 Fisher’s exact test to determine if there was a statistically significant association between diagnostic test outcome (true positive or false negative) and testing method (wet mount, zinc sulfate, sodium nitrate, or Sheather’s solution). The egg density data exhibited right-skewed distributions, so we log transformed the data to meet assumptions of parametric statistics before data analysis. We used a general linear model with diagnostic method (wet mount, sodium nitrate float, zinc sulfate float, Sheather’s float) as a fixed factor and fecal sample as a random effect to determine if there were significant differences between the mean egg densities observed among the four treatments using JMP v18 (SAS Institute, Cary, NC). A Tukey test was used to determine which methods significantly differed in mean observed egg density. For aged versus fresh fecal analysis, we used a Fisher’s exact test to determine if there was a difference in the number of false positive results between fresh and aged samples. We conducted a Passing-Bablok agreement analysis on the egg counts for fresh and dried subsamples followed a Bland Altman analysis in JMP v18. We also determined if the mean egg density differed between the fresh subsamples and the dried, aged subsample groups using a paired t-test.

To analyze the efficacy of our PCR assay, we separated our samples into groups by sample type (i.e., cloacal swabs, eggs in S. miliarius fecal samples, and pentastome adults and larvae) and by R. orientalis infection status (positive or negative). In this approach, each test result was treated independently based on the known infection status of the animal (confirmed by dissection or microscopy of fecal samples from S. miliarius), which allowed us to evaluate assay performance and sensitivity according to sample type. We calculated the sensitivity, specificity, positive predictive value, and negative predictive value for our R. orientalis assay across the three sample types (pentastomes, cloacal swabs, and fecal samples) using the confirmed positive and negative controls. Sensitivity was calculated as the proportion of animals that were correctly identified as infected by our assay (true positives) out of all animals known to be infected (the sum of true positives and false negatives). Specificity was calculated as the proportion of animals that were correctly identified by our assay as uninfected (true negatives) out of all animals that were uninfected (the sum of true negatives and false positives). The positive predictive value was calculated as the proportion of animals that were correctly identified by our assay as positive (true positives) out of all animals that tested positive, regardless of their true infection status. Finally, the negative predictive value was calculated as the proportion of animals that were correctly identified by our assay as uninfected (true negatives) out of all animals that tested negative, regardless of their true infection status. We quantified differences in detection rates between sample types by calculating Relative Risk for independent groups (i.e., cloacal swabs with paired fecal samples vs. non-paired cloacal swabs) and Odds Ratio for paired samples (i.e., fecal vs. cloacal swabs from the same individual). Confidence intervals were derived using the epitools package in R (Aragon, 2020; Posit, 2024; R Core Team, 2024).

For snakes of unknown infection status (i.e., cloacal swabs of live snakes and fecal samples with pentastome eggs that were not S. miliarius), we assessed the relationship of DNA concentration and amplification with a density plot. For positive control cloacal swabs, we assessed whether sample DNA concentration was associated with amplification probability using logistic regression. We only performed this analysis on positive control cloacal swabs given that almost all other positive control sample types amplified regardless of DNA concentration. In our analysis, we included a density plot to visualize the distribution of amplified samples relative to their DNA concentrations. The density plot was generated using ggplot2 in R (Wickham, 2016; R Core Team, 2024) and displays the likelihood of amplification across varying concentrations, helping to identify potential thresholds that may enhance amplification success.

To further validate our PCR assay, we tested the extraction and amplification protocols in a separate research laboratory at the University of Georgia (Athens, GA) using samples from nine snakes collected from Florida which were determined to have R. orientalis infections through postmortem evaluation and morphological pentastome identification. The snake species sampled included the black racer (n=1), common garter snake (n=1, T. sirtalis), S. miliarius (n=5), banded watersnake (n=1), and corn snake (n=1, Pantherophis guttatus). Adult pentastomes were collected and stored in 70% ethanol. Additionally, we collected fecal samples and cloaca swabs from each snake and stored samples at -80°C.

We cut adult pentastomes using sterile razor blades and forceps targeting a midbody section approximately 5 mm in length. Ethanol was evaporated off the cut section and extracted using the Qiagen DNeasy Tissue kit, following the manufacturer’s instructions (Qiagen, Hilden, Germany). We prepared fecal smear wet mounts (n=9) using the exact methods as above. We initially extracted fecal samples (n=6, samples with sufficient fecal content remaining following fecal smear wet mounts) using the Zymo Quick-DNA™ Fecal/Soil Microbe MiniPrep Kit following manufacturer’s instructions (Zymo Research, Irvine, California, USA). We repeated extractions when sufficient fecal material (n=6) remained, using the adapted fecal extraction protocol with the Qiagen DNeasy Tissue kit as detailed above. Cloacal swabs were extracted similarly to the above; however, were vortexed for one minute following the addition of 180 µL of Buffer ATL and 20 µl of Proteinase K and incubated at 56°C for one hour. Additionally, we performed an extra centrifugation step after the 3-minute centrifugation recommended by the manufacturer and used 100 μL of elution buffer. We used the same PCR and thermal cycler protocols however, we used two different thermal cyclers (Bio-rad DNAEngine; MJ Research PTC-200), and a different master mix. The master mix consisted of a 25 µL reaction volume with 5 µL of template DNA, 0.5 µL of 50 µM forward and reverse primers, 11 µL of molecular-grade water, 5 µL of Green GoTaq® Flexi Buffer (5x), 0.25 µL of dNTPS (80mM/dNTP), 2.5 µL of MgCl₂ (25mM), and 0.25 µL of GoTaq® Flexi DNA Polymerase (Promega Corporation, Madison, WI, USA). We Sanger sequenced all PCR products at the appropriate base pair length through Genewiz (Genewiz from Azenta Life Sciences, South Plainfield, New Jersey, USA).

For fecal sample analysis, all four diagnostic methods resulted in a high frequency of pentastome egg detection. The frequency of false negatives ranged from 0% for the wet mounts to 25% for the Sheather’s solution subsamples ( Table 1 ), and these differences were not statistically significant (Fisher’s exact test, P=0.071). The mean pentastome egg density observed was significantly affected by diagnostic method (F_3,57 = 32.53, P < 0.0001; Table 1 ). The mean number of eggs observed was significantly higher in the wet mount subsamples than in the three flotation solutions and the sodium nitrate and zinc sulfate floated subsamples had significantly higher means than Sheather’s solution subsamples (Tukey test, Table 1 ).

We observed only a single false negative flotation in the aged fecal subsamples and no false negative results among the fresh subsamples ( Table 2 ) resulting in no significant differences in the frequency of false negative results in the two subsample treatments (Fisher exact test, P=1.00). Agreement analysis for egg counts in the fresh and aged subsamples ( Supplementary File 3 ) indicated a significant bias between the two treatments, with the fresh subsample having higher egg counts than the dried subsample in 14 of 18 comparisons. The mean egg density in the fresh samples (1241.0 eggs/g) was significantly higher than the mean of the aged subsamples (mean = 366.6 eggs/g, t=3.30, df = 17, P=0.004).

We ran the Folmer et al. (1994) CO1 assay with six R. orientalis adults, which produced amplicons of 710 bp. DNA sequencing confirmed that all six morphologically identified samples were indeed R. orientalis with sequences being 100% identical to other R. orientalis CO1 sequences in GenBank (accession number MG559641). Additionally, the three individuals identified as P. crotali (n=3) were found to be 99–100% identical to P. crotali sequences in GenBank (accession numbers MG559654 and MG559653). We were unable to amplify R. orientalis DNA from the 16 cloacal swabs as well as the 11 blunt-tipped Raillietiella using this protocol, though we ran the PCR three times with positive controls (pentastome adults of other species) that always produced PCR bands of the expected size (710 bp).

For the 18S assay (Brookins et al., 2009), we successfully produced the target PCR amplicon size of ~425 bp in all cloacal swabs (n=16) and 14 fecal samples. Sequencing of these amplicons from cloacal swabs resulted in host DNA for all cases. For the fecal samples, sequencing results were mixed: three samples revealed host DNA (one of the positive control S. miliarius samples), while five samples yielded DNA from helminths (two of the positive control S. miliarius samples), despite confirming the presence of pentastomid eggs with wet mounts. Three fecal samples produced no amplicons, and three others returned sequences that were 100% identical to multiple pentastomid genera.

Notably, all 41 Raillietiellid individuals (R. orientalis and blunt tipped Raillietiella) amplified with the 18S assay produced amplicons of the target size (~425 bp). However, the returned sequences had 99–100% matches to multiple Raillietiella spp. in GenBank. Additionally, 19 of the R. orientalis individuals returned sequences identified as Reighardia spp. which infect seabirds as definitive hosts. For the other pentastome species, four K. coarctatus and three P. crotali samples produced amplicons of the target size (~425 bp). Of the K. coarctatus samples (n=4), two returned sequences with 99–100% identity to K. coarctatus (accession number MG559618), while the other two yielded sequences with 99–100% identity to multiple pentastome genera. Of the P. crotali samples, two yielded sequences with 99–100% identity to multiple pentastome genera, and one returned sequence with 99–100% identity only to P. crotali (accession number MG559598; Supplementary File 2 ).

Using our novel R. orientalis specific PCR assay, we recovered 100% positive predictive value across all positive controls, including R. orientalis adults, R. orientalis larvae, fecal samples with pentastomid eggs from S. miliarius, and cloacal swabs from snakes confirmed to harbor R. orientalis through dissection or inclusion in experimental infections ( Table 3 ). Additionally, all positive control samples had 100% specificity with zero false positives observed. We did not get amplification of adult samples from pentastome species other than R. orientalis, regardless of their DNA concentration ( Figure 1 ; Supplementary File 4 , Figure 1 ). Negative predictive values were 98% for adult and larval R. orientalis, with all samples amplifying except for one larva. Similarly, the negative predictive value for positive control fecal samples was 96%, with only one fecal sample from a confirmed R. orientalis-positive S. miliarius that did not amplify. With this one false negative, the sensitivity for detection in fecal samples was 97.6%, and with one false negative for individuals with known R. orientalis the sensitivity was 98.1%. In contrast, we recovered 47 false negatives from cloacal swab samples, translating to 67.4% negative predictive value and 21.7% sensitivity ( Table 3 ). We failed to amplify R. orientalis DNA in the six fecal samples that underwent fecal flotation.

Raillietiella orientalis individuals (adults and larvae) and fecal samples were more likely to amplify with our assay compared to cloacal swab samples ( Table 3 ). For positive control cloacal swab samples, we did not find a significant effect of DNA concentration on the probability of amplification (Estimate=0.00042, SE=0.0022, z=0.19, P=0.85). The odds ratio for amplification based on DNA concentration was 1.00 (95% CI: 0.99, 1.005) ( Figure 2 ). However, the density plot of amplification from cloacal swabs shows a left skew for samples that did not amplify ( Figure 3 ), suggesting that lower DNA concentrations may reduce the likelihood of amplification ( Supplementary File 4 , Figure 3 ). Similarly, fecal samples and cloacal swab samples from hosts with unknown status of R. orientalis infection also show that samples that did not amplify were skewed toward lower DNA concentrations ( Figure 4 ; Supplementary File 4 , Figure 2 ).

We collected fecal samples from 139 of the 296 live snakes with cloacal swabs collected. Therefore, these 139 snakes were known to have active fecal content in their cloaca at the time of swabbing. Of these paired samples (fecal and cloacal), 25.2% fecal samples (35/139, 95% CI: 0.18, 0.33) and 7.9% cloacal swabs (11/139, 95% CI: 0.04, 0.14) produced PCR bands of the expected size using our novel assay ( Supplementary File 4 , Table 1 ). Fecal samples were ~6 times more likely to test positive than their paired cloacal swabs (OR = 6.27; 95% CI: 2.79–14.06). Among the live snakes with non-paired cloacal swabs, 5.1% (8/157, 95% CI: 0.02, 0.1) yielded PCR bands of the expected size. Among cloacal swabs paired with positive fecal samples, 22.9% were also PCR positive (8/35; 95% CI: 0.10–0.40), compared to only 5.1% in the non-paired group. This represents a 4.94 times higher likelihood of detection via swabs from snakes with paired, positive fecal samples (95% CI: 2.20–11.08).

DNA sequencing with the R. orientalis-specific assay confirmed that all samples that produced PCR amplicons of the expected size ( Supplementary File 5 , Figure 4 ) produced CO1 sequence data specific to R. orientalis (all were 100% identical to other R. orientalis CO1 sequences in GenBank). Further, all sequences were 100% identical to each other, suggesting little sequence variation in this gene region, at least within the Florida population of R. orientalis (GenBank accession number PQ621873). All samples that amplified nontarget bands returned host (snake) DNA sequences. Further, we duplicated the reaction, gel, and sequencing for 95 positive control samples and received the sample results for all but one (98.9%). When compared with the results from the University of Georgia laboratory, results from our R. orientalis-specific assay were consistent with our initial findings, with 100% of R. orientalis adults (n=9), 44.4% of cloacal swabs (n=9), and 85.7% of fecal samples (n=7) producing the target amplicon, and all other samples producing no amplicons. Sequences from these samples were 100% identical to R. orientalis sequences in GenBank ( Table 4 ).