Morphological and molecular diagnosis of two new species of Trypanosoma Gruby, 1843 infecting South African cordylid lizards (Squamata: Cordylidae: Cordylinae), Trypanosoma (Squamatrypanum) ndumoensis n. sp. and Trypanosoma (Trypanosoma) tokoloshi n. sp.

Despite reptile trypanosomes forming a large group, the majority of species descriptions are data deficient, lacking key characteristic data and supporting molecular data. Reptile hosts show potential to facilitate transmission of zoonotic trypanosomiases and offer key information to understanding the genus of Trypanosoma. Several species of squamates from different localities in South Africa were screened molecularly and microscopically for trypanosomes in the present study. Based on the combination of morphological and molecular analyses, two new species of Trypanosoma, Trypanosoma (Squamatrypanum) ndumoensis n. sp. and Trypanosoma (Trypanosoma) tokoloshi n. sp., infecting South African cordylid lizards (Cordylidae: Cordylinae) are described in this study. The first molecular data for a South African reptile trypanosome is provided herewith.


I N T RODUC T ION
THE genus of Trypanosoma is a large group of unicellular, flagellate, extracellular obligate hemoparasites. These parasites have been described from all classes of vertebrates and from every continent. Several species are pathogenic causing severe, sometimes fatal, diseases in people and domestic animals. One such disease caused by these parasites and arguably the most well-known, is the infamous sleeping sickness, or Human African Trypanosomiasis (HAT). Monitor lizards have been shown to be naturally infected with Trypanosoma brucei, the pathogen responsible for causing HAT. These hosts were shown to experimentally transmit T. brucei, to laboratory-bred rats (Rattus norvegicus) and tsetse flies (Glossina morsitans) (Njagu, 1998), one of the main natural vectors responsible for the transmission of HAT to humans. The ability of some reptiles to potentially facilitate the transmission cycle of a human-infecting pathogen highlights the importance of researching reptile trypanosomes and gaining a better understanding of the group and their hosts.
Studying reptilian trypanosomes can be troublesome since infections are uncommon, typically with low parasitemia and infected hosts lack apparent symptoms. Effective species identification is hindered by a deficiency of nonpathogenic trypanosome molecular data and morphological plasticity. This is exacerbated by the fact that many species are unnamed or have poor descriptions. Subsequently, the validity of many taxonomic species is unclear, and many species have no additional records besides their original descriptions.
Currently, only 26 recognized species of squamate trypanosomes, shown in Table 1, have been reported from Africa (including Madagascar), generally with poor descriptions lacking characteristic data. Additionally, molecular data are available for only four species of African squamate trypanosomes on the NCBI GenBank (Trypanosoma varani (AJ005279), Trypanosoma sp. gecko (AJ620548), T. cf. varani (AB447493), and T. therezieni (AJ223571)). Despite lizard trypanosomes having the most reported species of the reptilian group (Viola et al., 2008), none have been described from South Africa. Of all the Squamata in South Africa, only two species of trypanosomes have been described from snakes, namely Trypanosoma sebae Fantham and Porter, 1950 and Trypanosoma psammophis Fantham and Porter, 1950. Cordylids (Cordylidae: Cordylinae) are a group of moderately sized, armored lizards and are the only lizard family restricted to mainland Africa. Only two species of Trypanosoma have been described from cordylid lizards, namely Trypanosoma cordyli Telford, 1995b, andTrypanosoma zonuri Telford, 1995b, from Cordylus tropidosternum and Cordylus cordylus senso lato in Tanzania (Telford, 1995b). A single species of Trypanosoma, T. betschi Brygoo, 1966a, has been described from Gerrhosauridae, the sister family of Cordylidae, from the host Zonosaurus madagascariensis (syn. Gerrhosaurus madagascariensis) from Madagascar (Telford, 2009). Currently, the trypanosomes of South African cordylid lizards remain unknown. The present study aimed to investigate and explore the diversity of South African squamate trypanosomes using microscopic and molecular screening and provide the first molecular data for South African reptile trypanosomes. The subgeneric classification system for the genus of Trypanosoma, proposed by Kostygov et al. (2021), was followed in the present study.

Collection of research specimens
Two Cordylus jonesii (Jones' girdled lizard) were found basking on tree trunks and collected by hand in Ndumo Game Reserve (KwaZulu Natal, South Africa) in March 2021. In total, fifteen Smaug depressus (Zoutpansberg girdled lizards) were collected from rock crevices by hand and using a noose from several sites on the Soutpansberg mountain (Lajuma Research Camp, Limpopo, South Africa) over a three-year period. Two S. depressus specimens were collected in November 2020, three in September 2021 and ten in September and October 2022.
Other squamates, including chameleons, geckoes, monitor lizards, skinks, and snakes, from several localities in South Africa were captured opportunistically by hand or with a noose (see Table S1). Animals were examined externally for parasites.
Blood was collected through a cardiac, femoral, or caudal venipuncture. Glass slide smears and blood samples were prepared using standard methods Netherlands et al. (2015). Blood smears were made onto a clean microscope slide, air-dried, and fixed with absolute methanol. Once dry, the slides were stained for roughly 20 min with a 10% solution of modified Giemsa (Sigma-Aldrich). Additional blood was stored in sterile 2-mL screw cap cryovials. Wet smears were made in the field of all ten S. depressus specimens captured in 2022, using a drop of blood underneath a cover slip on a clean microscope slide of all ten S. depressus specimens captured in 2022 and immediately screened using light microscopy. Additional blood samples were collected, and stained blood smear slides made using the aforementioned technique on two occasions after capture, from three of these S. depressus specimens.

Morphological characterization
Stained blood smears were screened for trypanosomes with a Zeiss AX10 microscope and images were captured using a Zeiss Axiocam 305 color at a magnification of 1600X. Morphometrics were measured according to a system adapted from Telford (1995a), Ferreira (2010), and Jordaan et al. (2023) using ImageJ version 1.52a (Schneider et al., 2012). Measurements of each morphological character were taken from 30 different specimens, when visible. Measurements include body length (BL); body width (BW) (measured at point of maximum width, excluding the undulating membrane); nucleus length (NL); nucleus width (NW); undulating membrane width (UMW); number of undulations (NU); kinetoplast length (KL); kinetoplast width (KW); mid-nucleus to anterior body end distance (MA); mid nucleus to posterior body end distance (MP); free flagellum length (F); mid-kinetoplast to anterior body end distance (KA); mid-kinetoplast to posterior body end distance (KP); mid-kinetoplast to mid-nucleus distance (KN). Derived measurements were nuclear index, NL/NW (NI); and body shape index, BL/BW (BI), are expressed as ratios. Nucleus position from the anterior end relative to body length, MA/BL (PN); and kinetoplast position from the anterior end relative to body length, KA/BL (PK), are expressed as percentages.

Mabuya quinquetaeniata)
Trachylepis albilabris (syn. using a NucleoSpin® Tissue DNA, RNA, and protein purification kit (Macharey-Nagel). Samples (see Table S1) were screened molecularly for trypanosomes by targeting two overlapping fragments of the 18S rRNA gene for amplification, using a nested PCR strategy adapted from McInnes et al. (2009) and Egan et al. (2020), following Jordaan et al. (2023). The primary PCR was conducted with the external primers SLF (5′-GCTTG TTT CAA GGA CTTAGC-3′) and S762.2 (5′-GACTT TTG CTT CCT CTAATG-3′). Two secondary PCRs were performed, using two different internal primer sets, namely B (5′-CGAAC AAC TGC CCT ATCAGC-3′) and I (5′-GACTA CAA TGG TCT CTAATC-3′); and S825 (5′-ACCGT TTC GGC TTT TGTTGG-3′) and SLIR (5′-ACATT GTA GTG CGC GTGTC-3′). The cycle conditions of the primary PCR included an initial denaturation step 95°C for 5 min, annealing step of 50°C for 2 min, and extension step of 72°C for 4 min. This was followed by 35 cycles containing a denaturation step of 94°C for 30 s, annealing step of 60°C for 30 s, and an extension step of 72°C for 2 min and 20 s. Lastly, a final denaturation step of 72°C for 7 min was performed. The conditions of both secondary PCR's consisted of an initial denaturation step of 95°C for 5 min, followed by 35 cycles of a denaturation step of 95°C for 30 s, annealing step of 57°C for 30 s, and an extension step of 72°C for 1 min. Thereafter, a final denaturation step of 72°C for 7 min was carried out.
Amplicons were electrophoresed with a 1% agarose gel to determine whether they were of the correct size (±900 bp) and desired quality for sequencing. Samples showing clear bands of the correct size were sent to Inqaba Biotech to be sequenced with the chaintermination method in both directions. Poor quality samples were discarded. Cultivation of these trypanosomes was not attempted.

Phylogenetic analysis
Geneious R9 (Kearse et al., 2012) was used to edit and trim the chromatograph sequences (received from Inqaba Biotec). A consensus sequence was created for each sample from the forward and reverse primer sequences of both B/I and S825/SLIR primer sets using the De Novo assembly method with Geneious R9 (Kearse et al., 2012).
A standard nucleotide BLAST™ (McEntyre & Ostell, 2013) search of the NCBI database was conducted using the final sequences to determine the percent identity matches of the samples. For the construction of a phylogenetic tree, comparable 18S rDNA sequences from the NCBI GenBank database were identified using BLAST™ (McEntyre & Ostell, 2013). Additionally, sequences were also adapted from the phylogenetic trees by Bernal and Pinto (2016) and Jordaan et al. (unpublished data). A multiple sequence alignment was constructed using the MUSCLE alignment method in Geneious R9 (Kearse et al., 2012) with default settings in FASTA format. The most appropriate nucleotide substitution model for the multiple sequence alignment was selected using jModelTest (Darriba et al., 2012;Guindon & Gascuel, 2003) with an Akaike information criterion (AIC) calculation. The general time-reversible (GTR) model (Tavaré, 1986) with inverse (+I) and gamma (+G) distribution, with a proportion of invariable sites of 0.3113 and gamma shape of 0.4154, was selected as the most suitable model. Phylogenetic trees using maximum likelihood (ML) analysis were constructed using RAxMLGUI 2.0 (Edler et al., 2021) with thorough bootstrap setting and 1000 bootstrap replicates. To perform a Bayesian inference (BI) analysis, the Markov Chain Monte Carlo (MCMC) algorithm with 10,000,000 generations was used in the program MrBayes (Huelsenbeck & Ronquist, 2001;Ronquist & Huelsenbeck, 2003), where every 100th generation was sampled. The first 25% of trees were discarded as burn-in. A model-corrected pairwise distance (p-distance) matrix was calculated in PAUP* v.4.0a169 (Swofford, 2003).

R E SU LT S
Of the 32 reptile specimens captured (see Table S1), ten of fifteen Smaug depressus (67%) and one of two Cordylus jonesii (50%) were found to be infected with trypanosomes after PCR screening. Microscopic screening of blood smears from an infected C. jonesii specimen revealed a positive infection, with a single trypomastigote morphotype and no dividing forms found. A positive infection was only observed in one of the infected Smaug depressus specimens (RE221002A1), after microscopic screening of the blood smears revealed only a single trypomastigote in one slide. A single live trypomastigote was observed in the wet smear of the same specimen, but in none of the other specimens. Positive infections of a single trypomastigote morphotype identical to the infection of S. depressus RE221002A1 were observed when microscopically screening the blood smear slides of the three resampled S. depressus specimens (RE221002A1, RE221001B4, and RE221002C1). An extremely low parasitemia was noted, with a single trypomastigote being observed only every second or third blood smear slide.
A small number of mites were found feeding externally on the body of the noninfected cordylid specimen, preliminarily identified as Trombiculidae larvae (Castro et al., 2019). Ectoparasitic mites preliminarily identified as Trombiculidae larvae, Ophyionyssus-like sp., and Pterygosomidae mites were observed on several of the infected S. depressus specimens. It is unknown whether or not these mites are potential vectors for trypanosomes. Representative DNA sequence: The sequence data specifically associated with T. ndumoensis n. sp. (upon which the present biological description is based) has been submitted to GenBank and are as follows: Nuclear 18S rDNA (nu 18S) partial sequence OQ297738.
Etymology: The species epithet is in the form of a noun in apposition, constructed from the name of the type locality, Ndumo Game Reserve, where the type host specimen was collected.
Site of infection: Peripheral blood. Vector: Unknown.
Remarks: This trypanosome has a broad mid-body and short tapering ends, with several granular vacuoles often visible ( Figure 1A-L). The cytoplasm stains a shade of light purple, fading in intensity until almost transparent toward the undulating membrane. When fixed and stained smears were examined under a microscope, the body was often curled in a U-shape, with the anterior and posterior ends sometimes touching. The nucleus and kinetoplast were observed to be closely associated and were positioned, on average, two thirds from the anterior body end. The nucleus was observed to stain pink and the kinetoplast a dark purple. A relatively large, almost transparent undulating membrane and long transparent flagellum were also notable.
Trypanosoma (Squamatrypanum) ndumoensis n. sp. appears closest in morphology to T. zonuri Telford, 1995b, but when compared to the three least variable characters of T. zonuri (BL, BW, and KA), according to Telford (1995b), T. ndumoensis n. sp. is clearly a separate T A B L E 2 Morphometric measurements of Trypanosoma ndumoensis n. sp. and Trypanosoma tokoloshi n. sp. from the present study. 69 μm) appears larger than that of T. zonuri when compared to its illustration with the scale accounted for, however, a measurement was not included in the original description for comparison. The sequence of T. ndumoensis n. sp. is well nested within the Squamatrypanum clade of the phylogeny, forming a monophyly with Trypanosoma sp. 109 (LC471399) and Trypanosoma sp. gecko (AJ620548), and sister to Trypanosoma varani Wenyon, 1908 (AJ005279) and T. cf. varani (AB447493). While no morphological data was provided for its two closest relatives, it does share morphological characters with T. varani (AJ005279) and T. cf. varani (AB447493) (Sato et al., 2009). The body shape of T. varani is similarly described as U-shaped by the original author (Wenyon, 1908). Although only basic measurements of T. varani were included in the original description, it can be deduced by using the scale in the illustration (Plate XIII, figures 11-13 in Wenyon, 1908) that the nucleus was roughly 3.24-4.97 μm long and 3.18-3.67 μm wide. Despite being an approximate estimation, this range is more than double that of T. ndumoensis n. sp. (1.2-2.16 × 0.92-1.82 μm). Trypanosoma cf. varani has a similar body width (9.8 μm) to T. ndumoensis n. sp. but is significantly longer (51.8 μm) with a shorter free flagellum (4.2 μm). A distance of 4.4 μm separates the kinetoplast and nucleus of T. cf. varani, unlike the close association observed for T. ndumoensis n. sp. Posterior to kinetoplast measurements of T. varani (15 μm) and T. cf. varani (16.4 μm) are greater than the MP distance (an analogous measurement) of T. ndumoensis n. sp. (11.75 μm).
Although closely related, the host species of Trypanosoma ndumoensis n. sp. and T. zonuri are separated geographically by the arid corridor (Freitas et al., 2018) and Cordylus jonesii, the host of T. ndumoensis n. sp., was shown by Greenbaum et al. (2012) to have a separate distribution from the Tanzanian species of Cordylus referred to in the description of Telford (1995b) as C. cordylus senso lato, the host of T. zonuri. Therefore, due to the difference in host species, geography, morphology, and with supporting molecular data, Trypanosoma ndumoensis n. sp. is shown to be distinct species of Trypanosoma, from a previously unreported host species.
Subgenus: Trypanosoma Gruby, 1843 emend. Votýpka and Kostygov, 2021 Description of Trypanosoma tokoloshi Jordaan, van As and Netherlands n. sp. Etymology: The species epithet is constructed from the Sepedi name, one of the most widely spoken languages in Limpopo from where this parasite originates, for a mischievous creature in South African folklore which has an aversion to being seen (noun in masculine genitive case form). Thus, this parasite was named after the Tokoloshe for the difficulty of finding this parasite during microscopic screening, due to its extremely low parasitemia.
Site of infection: Peripheral blood. Vector: Unknown. Stages in Vector: Unknown.
The host of T. tokoloshi n. sp., Smaug depressus, has a distribution restricted to the Soutpansberg mountain range and surrounding areas in Limpopo, South Africa, in addition to also being separated from host species of T. cordyli, C. tropidosternum, by the arid corridor (Freitas et al., 2018). Consequently, due to the morphological F I G U R E 2 Trypanosoma tokoloshi n. sp. plate. (A-L) Normal trypomastigote forms in the blood. Arrowheads show kinetoplast (A, B, and K); arrows show flagellum (A, B, and F) and undulating membranes (C, K, and L); nuclei are indicated by "n" (B, J, and K). Scale bar is 10 μm. characteristic differences, different host species, and geographic distribution, Trypanosoma tokoloshi n. sp. is a distinct species. as one of the nested PCR's was unsuccessful. There was a 0.14% intraspecific divergence between the sequences OQ297740 and OQ297739 (Table S2), which was due to OQ297740 differing by five base pairs, three of which were called as ambiguous nucleotide base pairs during editing of the chromatograph sequences due to a slightly lower quality sequence. Sequencing of S. depressus sample RE221002A1 was unsuccessful, although it was shown to have a positive trypanosome infection when screened molecularly with a nested PCR using trypanosome primers. An nBLAST query of these sequences returned results showing a close relation (> 99.67%) of T. ndumoensis n. sp. to trypanosome sequences (Accession numbers LC471393 to LC471399) isolated from Italian sand flies, Sergentomyia minuta, (Phlebotominae) in the study of Abbate et al. (2020). As these sand fly sequences were identical, they were represented by the single sequence of Trypanosoma sp. 109 (LC471399) in the phylogeny. It is important to note that the sequences of Abbate et al. (2020) were significantly shorter in length (834-917 bp) than the sequence of T. ndumoensis n. sp. from this study (1499 bp), which could lead to a lower accuracy of the comparison. The second closest match was 99.13% with the sequence of Trypanosoma sp. "gecko" (AJ620548), isolated from a Senegalese Tarentola annularis specimen, a species of gecko native to northern Africa. An nBLAST query of the T. tokoloshi n. sp. sequences returned several matches below 98% with unnamed anuran trypanosomes from South America and Europe, with the closest match being 98.05% with the sequence of Trypanosoma sp. 858 (EU021228) isolated from the South American toad, Rhinella icterica.

Molecular and phylogenetic analysis
The topologies of the maximum likelihood (ML) and Bayesian Inference (BI) phylogenies constructed using 18S rDNA trypanosome sequences were identical and were thus combined into a consensus phylogram in Figure 3. The phylogenetic relationships were consistent with described trypanosomatid phylogenetic relationships. Clades in the phylogram were well-supported, with the majority of bootstrap and nodal support values above 70% and 0.80, respectively.
The sequence of Trypanosoma ndumoensis n. sp. is placed within the Squamatrypanum clade (subgenus Squamatrypanum Votýpka and Kostygov, 2021) in a monophyletic subclade closest to the European sandfly sequence, Trypanosoma sp. 109 and the West African gecko trypanosome sequence, Trypanosoma sp. "gecko." Although, it should be taken into consideration that the sandfly sequence was significantly shorter in length and might, therefore, not necessarily represent an accurate relationship.
The ten sequences of T. tokoloshi n. sp. isolated from the Smaug depressus specimens had an identical genotype and were thus represented by two of the sequences, OQ297739 and OQ297740. The sequences are placed in a new monophyletic subclade nested within the anuran trypanosome clade (subgenus Trypanosoma Gruby, 1843 emend. Votýpka and Kostygov, 2021), closest to two unnamed anuran trypanosome species (MH424271 and MH424274) infecting Ukrainian Pelophylax ridibundus (Spodareva et al., 2018).

Morphological characterization
Reptile trypanosomes are typically found as trypomastigote forms in the blood of their hosts and undergo different morphological transformation stages within the vector (Telford, 2009). Generally, only one trypomastigote form is observed in the squamate host, however, some species have been described as polymorphic, such as T. martini (syn. T. mabuiae), in spite of this not being definitively validated. Although it is difficult to distinguish species of Trypanosoma based on morphology alone, studies conducted by Telford (1995aTelford ( , 1996Telford ( , 2009 indicate comparative morphological analysis is a valid method of distinguishing saurian trypanosomes, despite their morphological variability. Among fourteen saurian trypanosome species, the body length and width were reported to be the most consistent measurements between species (Telford, 2009). Furthermore, additional morphological characteristics which aided in distinguishing species were differences in staining properties, the average position of the nucleus and kinetoplast relative to the body, the nuclear index, and body shape index (Telford, 2009). Differences in staining color could potentially be explained by differing concentrations of lipid compounds in the cytoplasm between species, as a result of the Giemsa stain being water-buffered. This was observed with the lighter-staining, vacuolated cytoplasm of Trypanosoma ndumoensis n. sp. in contrast to the darkerstaining, less vacuolated cytoplasm of T. tokoloshi n. sp.
During initial sampling, no trypomastigotes were observed during light microscopy screening of the blood smear slides of Smaug depressus, which were shown to be positive when screened with trypanosome primers using a nested PCR and gel electrophoresis. The absence of trypanosomes in the peripheral blood, despite a known positive infection, has been reported by other researchers previously. Telford (1995b) kept captive cordylid lizards known to be infected with trypanosomes over several years and periodically drew and screened their blood, noting on several occasions that no trypanosomes were present. He stated: "Parasites were often absent on [blood] slides taken at irregular intervals [over a 4-year period]." Viola et al. (2009) were also unsuccessful in detecting trypomastigotes of the South American snake trypanosome, Trypanosoma serpentis, in blood smear slides with light microscopy screening. It was hypothesized by Viola et al. (2009) that very low parasitemia in infected hosts was the cause thereof. Ferreira et al. (2017) also noted the absence of trypomastigote forms in the blood smears slides of South American marsupials, Monodelphis domestica, infected with Trypanosoma gennarii. Lastly, Fermino et al. (2019) were unable to detect blood trypomastigotes of T. kaiowa in the blood smears of infected crocodilians but were able to identify developmental forms in the invertebrate vectors with molecular confirmation. Viola et al. (2009), Ferreira et al. (2017, and Fermino et al. (2019) utilized culturing methods to study the morphology of the aforementioned species of Trypanosoma, however, it has been shown that the morphology of cultivated trypanosomes can differ to the morphotypes observed in natural infections (Diamond, 1965). A wetmount slide preparation technique was implemented in the present study as a method for more successful detection of trypanosomes when screening in the field; however, it was only as effective as the standard technique of screening stained blood smear slides. It appears that, as a result of the extremely low parasitemia and, in spite of thorough microscopic screening of the blood smear slides, the infections are often not apparent. Ultimately, we recommend PCR screening using trypanosomespecific primers for trypanosomes.

Molecular and phylogenetic analysis
Although it is unexpected that T. tokoloshi n. sp. clades with anuran trypanosomes, other lizard trypanosomes have been previously reported to clade in the anuran group as well-such as T. therezieni, isolated from a chameleon host (Haag et al., 1998). Additionally, Trypanosoma sp. DNA was reported from geckoes in Thailand, which claded with anuran trypanosomes in the phylogeny (Toontong et al., 2022). It is unclear whether the findings reported by Toontong et al. (2022) were due to an incidental infection of geckos which fed upon sandflies infected with anuran trypanosomes; if it was a natural infection of a reptile trypanosome that merely clades with the anuran trypanosomes in the phylogeny; or if it is a species of trypanosome that infects both reptiles and anurans naturally. Perhaps, instead, these reptile trypanosomes clade with the anuran trypanosomes because of the lack of phylogenetic coverage due to the small amount of lizard/anuran sequences available or have split evolutionarily from a separate ancestral lineage to the other reptile trypanosomes.
In the present study, the presence of trypanosome infections in nine Smaug depressus lizards from several localities suggests that it is unlikely to be an incidental infection of an amphibian trypanosome in reptile hosts, but rather a reptile trypanosome that clades with anuran trypanosomes or a trypanosome which is able to infect both classes of host species. The current evidence of three reptile trypanosomes being placed in separate anuran clades possibly indicates that there are multiple points of divergence in the ancestral lineage of reptileinfecting trypanosomes, with some clades having diverged more recently from a shared common ancestor with anurans. In contrast to crocodilian and varanid trypanosomes which appear to have diverged at an earlier point in the ancestral lineage. It is hypothesized that ancestral trypanosomatids were initially monoxenous parasites infecting insects (likely dipterans) and later became dixenous due to hematophagous invertebrate hosts (Spodareva et al., 2018). Based on the phylogenetic evidence, Spodareva et al. (2018) suggest, that terrestrial trypanosomes originated from amphibian trypanosomes, as is shown by closely related infections in turtles, platypuses, fishes, and chameleon hosts. The lizard trypanosome in the present study, T. tokoloshi n. sp., and the trypanosomes infecting geckos reported by Toontong et al. (2022), which both clade within amphibian trypanosome groups could potentially be evidence in support of this hypothesis.

CONC LUSION
Molecular and morphological evidence indicates that two distinct species were observed in two reptile host species during this study, Cordylus jonesii and Smaug depressus. Subsequently, two new species of Trypanosoma are presented in the present study with supporting morphological descriptions and molecular data. Molecular screening was shown to be more reliable than screening with light microscopy, due to the extremely low parasitemia of these lizard trypanosome infections. Phylogenetic comparisons with other trypanosomes of a similar evolutionary history, suggest the possibility of a bloodsucking arthropod, such as a phlebotomine sandfly, as the vector, however, potential vectors for lizard trypanosomes need to be investigated in future studies.